PRT obsoleted by robocars

[Brad Templeton]

In response to some discussion on this group of my robocar articles in
relation to PRT, I wrote an expanded section on why I make the rather
bold claim that a not-yet practical technology like robocars could
make us declare a never-really-implemented technology like PRT to be
on the verge of becoming obsolete.

It's at http://www.templetons.com/brad/robocars/prt.html

I started out as a fan of PRTs basic ideas -- personal vehicles,
little waiting and non-stop station to station transportation which
would give riders something much closer to what they get from cars,
while keeping many of the benefits of transit (no driving, no parking,
limited congestion.) I even dreamed of hybrid PRTs that left the
track and were driven the last mile on city streets. But then,
thanks to DARPA, robocars started moving from SF to reality and could
offer more than PRT ever promised without requiring any new
infrastructure.

So how to you compare a technology that's not here yet, with a
technology that's seemed possible for 40 years but never realized?

The answer for me came by looking at who pays for it. Robocars can be
privately owned and funded, with very minimal government involvement.
As their abilities improve with Moore's law, they will be bought by
rich early adopters, then bought again next year by the same early
adopters. This approach drives incredible, unbeatable innovation.
PRT, like all transit, has to be bought by risk-averse municipal
transit planners with decade long time horizons and huge budgets. It
is almost a rule that what they buy will be obsolete by the time they
break ground, let alone by the time it's in operation.

Anything that brings Moore's law, individual purchase, early adopters
and highly competitive markets to transportation is sure to win. I
don't think there's even a contest. PRT's few remaining advantages
(better efficiency through electrified rails and less congestion on
custom-built ROW) barely blip the radar. Thus the bold statement.
Give it a read for more details.

[Dennis Manning]

Brad:

 

At your web site the last paragraph was:

 

• Finally, if traffic is very thick that this 20-fold increase in capacity is unable to handle it, we can have robo-buses, or trains using legacy rails. People wishing to travel a very heavily used route will have their robocar take them to an appointed place and time where they switch to a bus, which then travels nonstop along the route. At the end, it stops in a small lot where many single-person robotaxis are waiting to take them the rest of their way. This should allow a very large increase in the capacity of the roads, at the cost of having buses that are used only at these peak times -- still cheaper than building new right-of-way.

 

 

You didn't say much about public transit. This paragraph is sort of confusing. Can you expand a bit on how robocars or robobuses might help the 30% or so of the population that can't drive?

 

My knee jerk reaction is why not both PRT and robocars? So far as I can see there will always be two markets - private and public.

 

Dennis

[Jerry Roane]

"(better efficiency through electrified rails and less congestion on
custom-built ROW) barely blip the radar."

Brad

You mean $4.00 gasoline has not blipped the radar. 

In a solar version we do not use ANY FORIEGN OIL or fossil fuel.  It is through the guideway energy storage and sipping that this is even possible.  That is the radar.  Our system does not use rails but we do supply enough energy to go fast under the solar energy budget.  Our system patents describe how the robotic control is accomplished in the city.  How is the trick.  If you heard the latest energy industry commercial they say we have enough domestic oil for 60 years.  Do they even listen to their own noise?  What about year 61?  Then what? --- Robo cars? 

What sand do you have your head stuck in?  Did you hear the angry chant, "Drill Drill Drill" at the convention?  Robo cars are not going to build more highway lanes alone so unless more lanes are built robo or hot-headed drivers are still going to be congested.  What makes you think robo control will increase the per lane capacity past say California highway drivers?  I do like the bold approach but it is most likely misdirected.  Robotronics can help the chicken drivers but it won't take the lumbering trucks out of the way.  Any grade change on a highway through a city causes these trucks to start the ripple effect in the traffic flow.  Individual robo cars will have the exact same inductive control but with only a slightly faster response time after the compute time of the expert software.  If the cars are on a master control system then all cars will need to participate and that will not happen.  You will need to count on a significant (till people die off) overlap of robo and driver cars.  This requires that the robo car be inductive in the response.  A robo car that does not have superior braking like TriTrack and other guideway direct braking systems will crash just like every other car with 1.1 Max G force brakes.  Rubber tires are a problem that robo alone does not fix.  If you combine robo with guideway then I can agree with some part of your assertion.  The Moore's law thing is incorrect of course because it is not a law--- more of a guideline.  It cannot predict the future any more than a rabbit's foot but the economic driver is a valid point you make.  On a guideway robo driving is reduced to --- stop or continue.  Any computer could pull off that control.  Even ones from the 1970s.  Also being an electrical engineer I can tell you that the control computers of the 1970s although not 9 digit accurate were faster at computation than the fastest digital computer of today.  It was a different approach but very fast.  I seem to remember we put men on the moon using these fast computing electronic approaches to control. 

The federal highway fund is bankrupt.  Do you mean that a fully bankrupt highway system can supply the needs of the population in 2008+ when they didn't supply the needs of 2001?  It is easy to take two data points and extrapolate to infinity using X-cubed terms but reality is less exciting usually.  I am sure we will have better video games/movies but the rest will lag considerably.  We are neutral on how the hardware is operated.  If a city wants to run the hardware pure PRT, half PRT or pure DM that is their choice.  Robo cars will enhance the number and depth of the choices but I doubt it will end any movement.  We think that robo cars are eventually going to be part of the freight mix first but these cars need a path on which to travel no matter what is actuating the steering wheel and pedal. 

Jerry Roane

[Brad Templeton]

On Sep 6, 2:51 pm, "Jerry Roane" <jerry.ro...@gmail.com> wrote:
> "(better efficiency through electrified rails and less congestion on
> custom-built ROW) barely blip the radar."
>
> Brad
>
> You mean $4.00 gasoline has not blipped the radar.
>

> In a solar version we do not use* ANY FORIEGN OIL* or fossil fuel.

There is no such thing as "a solar version." We have an electrical
grid
with a mix of sources. Currently 50% coal, 20% natural gas, 20% nuke
and 10% hydro, with a little solar/renewable in the noise. If you
build more
solar power, great, but you would be crazy if you didn't simply hook
it up to
the grid. Solar powered transit is just plain silly. Really,
really silly.

So by adding this solar power to the grid, you would be greening the
grid a little,
which is great, but it really says nothing about your transportation
system
per se,. Mainly, electric transportation (electric car or train or
trolley) has
the grid's footprint, and gasoline powered transportation has that
footprint.

Do you have figures showing what efficiency you expect from a guideway
powered PRT pod compared to a battery powered ultralight electric
vehicle?
What about flywheel or compressed air?

Based on Tesla's figures, they show about a 15% loss due to charging
and battery issues.
Do you have some other figures? Not that I want to lose 15% but it's
pretty minor,
especially when you consider that electric cars can charge "smart"
from the grid,
taking power at off-peak times when it is cheaper, while live-driven
vehicles must take
power when they move, including at afternoon rush hour which is
usually the grid peak.

> What sand do you have your head stuck in?

Whoops. That sort of talk ends any debate with me. If you would care
to repost your
thoughts without this sort of statement in them, and pledge never to
make such statements
again in your apology, I will read them.

[Brad Templeton]

On Sep 6, 2:42 pm, "Dennis Manning" <john.manni...@comcast.net> wrote:
> Brad:
>
> At your web site the last paragraph was:
>

> a.. Finally, if traffic is very thick that this 20-fold increase in capacity is unable to handle it, we can have robo-buses, or trains using legacy rails. People wishing to travel a very heavily used route will have their robocar take them to an appointed place and time where they switch to a bus, which then travels nonstop along the route. At the end, it stops in a small lot where many single-person robotaxis are waiting to take them the rest of their way. This should allow a very large increase in the capacity of the roads, at the cost of having buses that are used only at these peak times -- still cheaper than building new right-of-way.

>
> You didn't say much about public transit. This paragraph is sort of confusing. Can you expand a bit on how robocars or robobuses might help the 30% or so of the population that can't drive?

It is my expectation that robocars would replace public transit over
time, as they can be superior by every metric, including energy
efficiency. I don't understand
your question about the population who can't drive -- they should love
robocars more than anybody, surely.

[Dennis Manning]

----- Original Message -----
From: "Brad Templeton" <bra...@gmail.com>
To: "transport-innovators" <transport-...@googlegroups.com>

Are you suggesting robocars require no drivers license and every one can
afford a robocar thereby eliminating public transit? Surely you jest.

BTW if a computer is controlling the car who is liable in an accident?

Dennis

[Jay Andress]

Brad,

 

  Below I have reposted my comments from eight days ago. I think you are being too idealistic about Robocar. I don't think it will be safe at all (are you aware of the problems with how difficult it was for the Honda Robot to walk down a sidewalk...image the problems with cars). I don't understand how vehicles are suppose to be readily available, fully charged yet your energy numbers seem to be based on only trips when vehicles are occupied. Where are these vehicles stored? Where are these vehicles charged? How does Robocar handle large crowds...like 30,000 people exiting a city every afternoon. Finally how much do these vehicles cost?

  It is interesting that you have looked at hybrid PRT and Robocars...I think that has some potential.

  Below is my earlier post:

 

 

Robocar myths....

 

At first glance the idea of using robotic cars is appealing but there are serious questions about it, that have been ignored...
  1) I'm not sure they will be safer...which appears to be the main argument. When I think of most serious car accidents they are usually caused by something going terribly wrong...usually between a truck and a car (the last four people killed in car accidents in Cincinnati were killed by trucks). In every situation the victims had no chance of avoidance...how is Robotic Car (RC) solving that...absolutely nothing. I'm not sure RC will do better with pedestrians. How many DARPA officials walked in front of the vehicles? My guess....zero. Will (RC) be able to stop if a child on a bicycle rides in front of it? Will RC come to a stop for every person on a bike or change lanes creating a potential crash? Will RC steer around the dog and hit the mother walking the child? Will RC be able to discern that the pedestrian walking down the street is not looking in the direction of the vehicle and may step out into the road? I think safety claims are difficult with new technology (with dual mode for example we will prevent all head-on crashes but what happens at 100 mph at a switch?)
  2) Vehicles that mysteriously appear out of nowhere within seconds (this also applies to PRT) is a myth. Those vehicles had to be sitting or running empty nearby. Where are the parking lots to store those waiting vehicles? If I have 30,000 people exiting a major urban area at 4:30 pm where are all the vehicles at 4:15 pm? If I have 80,000 people leaving a stadium, where are the 50,000 vehicles suppose to be waiting? Some would argue that because each vehicle can make multiple trips it would require fewer vehicles...that is not going to cut it...people will not wait for the vehicle to return... a wait of 30 minutes or more.
  3) The energy calculations are all wrong (and therefore the environmental impact). There is a trade-off between robotic cars readily available and energy. This is a serious flaw in Brad's argument for robotic cars...he only includes the energy numbers based on travel when the passengers are in the vehicle...yet these vehicles would have to be running almost constantly (and burning fuel) to be as available as he claims. Also he says that they will automatically go to recharging stations yet does not include any energy in his calculations to get there. Where are all those robotic cars parked while they are recharging? Not in my neighborhood.
 
  I think PRT with robotic cars makes more sense. It provides mass transit type service in crowded urban areas and helps congestion by being separated from regular traffic. The PRT rail can recharge batteries during transit. The PRT rail does not have to go into every neighborhood since vehicles would disconnect from the rail and travel automatically to a neighborhood (for safety reasons I think that robotic cars when off the rail, would have to run in designated lanes, perhaps in the middle of the street to keep them away from pedestrians).  In heavily congested areas where it is necessary to move people quickly, there is a track that can move people with only a few second spacing. IMO a PRT system like ULTRA with robotic cars has great appeal for an entire urban area (although my first choice will always be Dual Mode)
 
                                                                            Jay

[eph]

A theoretical robocar will be nice in the future when the kinks are
all worked out and they become affordable, until then, Billions of
dollars are being wasted on public transit systems that could be so
much better and cheaper with PRT/DM solutions that can work now.

F.

[Brad Templeton]

On Sep 6, 3:39 pm, "Jay Andress" <andress....@gmail.com> wrote:
> Brad,
>
>   Below I have reposted my comments from eight days ago. I think you are
> being too idealistic about Robocar. I don't think it will be safe at all
> (are you aware of the problems with how difficult it was for the Honda Robot
> to walk down a sidewalk...image the problems with cars). I don't understand
> how vehicles are suppose to be readily available, fully charged yet your
> energy numbers seem to be based on only trips when vehicles are occupied.

Well, you may think of this as a definitional argument, but of course
I don't believe they
will be deployed until we are convinced they are safe. What I write
about is, given the
assumption that we can build a safe robocar, what does does that mean
for transportation.

If you don't think we can build such a car, then of course you will
say it means nothing.
However, evidence is building that it can be done, certainly with the
will. Robocars have
a big leg up over humans at being safer. 80% of accidents are caused
by inattention, which
robocars would not suffer from. They will always be looking in every
direction, something a
human can't do. And they will react much more quickly.

That said, of course when they focus their attention they must
correctly identify what they
are looking at. Nobody would pretend that this is not a hard
problem. But it seems today to
be one that is within our grasp. But as a future technology, we can't
prove that until we
do it.

> Where are these vehicles stored? Where are these vehicles charged? How does
> Robocar handle large crowds...like 30,000 people exiting a city every
> afternoon. Finally how much do these vehicles cost?

The answers to these questions change over time. They would probably
be
stored on the streets, and in existing parking spaces. They can be
stored blocking
driveways, and packed at 2-3 times the density today's cars get in
parking lots.

They can charge (if electric) anywhere there is power. They go to the
power, it
doesn't have to be everywhere. I would predict simple cheap poles all
over the
city that a robocar can pull up next to, authorize payment, and tap
onto. But there are
other methods -- including many that are not electricity. New power
schemes can
be experimented with at will because the cars go to the where the fuel
is on their
own.

For the question of large crowds, did you follow the link in my main
post?

Cost? Initially quite expensive and limited to the rich. Then,
thanks to Moore's
law, cheaper then regular cars. Then, once most of the HDVs are off
the road so
the cars don't need expensive safety systems, quite a bit cheaper than
regular cars.

Simple electric city cars should be quite cheap if mass produced.
It's just motors and
wheels, no drivetrains, transmissions, etc. The computer part
starts expensive and becomes
$100 like most other computer parts. I could easily see a one
person, 3-wheeled, fiberglass
shelled vehicle with 1kwh of lead-acid battery costing under $2K.

>   It is interesting that you have looked at hybrid PRT and Robocars...I
> think that has some potential.

Sort of. If you don't have robocars, then PRT is only minimally
marketable.
In fact, based on past history, PRT on its own is a complete failure
-- nobody
has sold a real PRT system, just a couple of toy systems.

If you do have Robocars, why would you bother with PRT?

So I no longer think the hybrids are so interesting.

> argument. When I think of most serious car accidents they are usually caused
> by something going terribly wrong...usually between a truck and a car (the

No, that's not correct. Figures I have read cite that 80% of
accidents are caused
by the driver not paying attention (in many cases due to being tired
or drunk.)

Robocars will never not pay attention, at least.

> better with pedestrians. How many DARPA officials walked in front of the
> vehicles?

With the first prototypes? Why would you expect them to? Of course
before
they are on the streets they will.

My guess....zero. Will (RC) be able to stop if a child on a
> bicycle rides in front of it?

Of course! We would hardly accept them if they did not. Though
probably not stop, but turn to avoid, stopping only if there is no
other option.

> Will RC come to a stop for every person on a
> bike or change lanes creating a potential crash?

Why would it do something to create a potential crash?

>Will RC steer around the
> dog and hit the mother walking the child?

That would be pretty stupid, I think.

> Will RC be able to discern that
> the pedestrian walking down the street is not looking in the direction of
> the vehicle and may step out into the road?

That's a more interesting question. My expectation is they will
always leave an escape gap, unless we decide that in very high usage
roads we want them to pack so densely there is no escape gap. I
don't know if we want to do that right away.

But normally the vehicle would drive so as to always have a way out if
something unexpected happens, like a pedestrian stepping
on the road. Of course robot reaction times are far better than
human.

You did read the bit about the school of fish?

I think safety claims are
> difficult with new technology (with dual mode for example we will prevent
> all head-on crashes but what happens at 100 mph at a switch?)
>   2) Vehicles that mysteriously appear out of nowhere within seconds (this
> also applies to PRT) is a myth. Those vehicles had to be sitting or running
> empty nearby. Where are the parking lots to store those waiting vehicles?

If vehicles are in use, then they are not parked. If they are not in
use, they
inherently need less road, so more road can be allocated to parking.
Robocars
can park in front of driveways, can double and triple park.

If we should ever want so many vehicles that is. Today there's 2 cars
in every
garage and a chicken in every pot. On-demand vehicle within a minute
can be
done with vastly fewer vehicles than we have now, if we wish it. It's
really a question
of what we want to pay for.

> I have 30,000 people exiting a major urban area at 4:30 pm where are all the
> vehicles at 4:15 pm? If I have 80,000 people leaving a stadium, where are
> the 50,000 vehicles suppose to be waiting? Some would argue that because
> each vehicle can make multiple trips it would require fewer vehicles...that
> is not going to cut it...people will not wait for the vehicle to return... a
> wait of 30 minutes or more.

There are situations where we might use shared vehicles (bus/train/
van.) That's always
available as a fallback to any challenge like this. However, I
suspect we don't need to
go to that fallback. It's an interesting challenge. There is *so*
much parking available
if your challenge is to park robots that will move on command and
which are just 4' wide
and 10' long. You can store 1,000 of them in an acre! 50,000 could
go into a 50 acre
which is I think just a small fraction of the typical parking lot
found at today's sports stadiums.

However, you don't need that as you can store them right on the roads
themselves. When
the roads are busy, you don't need to store them. When the roads are
not in use, you only need
to keep one or two lanes open. (Most roads, including their current
parking lanes, are about 40 to 50 feet
wide.)

>   3) The energy calculations are all wrong (and therefore the environmental
> impact). There is a trade-off between robotic cars readily available and
> energy. This is a serious flaw in Brad's argument for robotic cars...he only
> includes the energy numbers based on travel when the passengers are in the
> vehicle...yet these vehicles would have to be running almost constantly (and
> burning fuel) to be as available as he claims. Also he says that they will
> automatically go to recharging stations yet does not include any energy in
> his calculations to get there. Where are all those robotic cars parked while
> they are recharging? Not in my neighborhood.

Why not? If you want to summon them in your neighbourhood you have
to accept
them there. You are correct, we do not yet have figures for how much
idle travel is
needed. Note that I expect vacant energy use to be a fair bit below
occupied energy use, as
the vehicle may not weigh much more than the passenger in the final
all-robocar
world where ultralights are present. But even with equal unoccupied
travel the numbers
are still pretty good, and I don't think we have to get close to that.

PRT has the same issue of course, but must build spur-lines for
storage. Robocars can
store wherever there is pavement.

>
>   I think PRT with robotic cars makes more sense. It provides mass transit
> type service in crowded urban areas and helps congestion by being separated
> from regular traffic. The PRT rail can recharge batteries during transit.
> The PRT rail does not have to go into every neighborhood since vehicles
> would disconnect from the rail and travel automatically to a neighborhood
> (for safety reasons I think that robotic cars when off the rail, would have
> to run in designated lanes, perhaps in the middle of the street to keep them
> away from pedestrians).  In heavily congested areas where it is necessary to
> move people quickly, there is a track that can move people with only a few
> second spacing. IMO a PRT system like ULTRA with robotic cars has great
> appeal for an entire urban area (although my first choice will always be
> Dual Mode)

Again, once we have the robotic guiding, why have the PRT? The truth
is "special lanes"
is a short-duration solution. The period where a robocar can operate
safely in its
special lane but is not safe to operate outside it will be only a
small number of years.
Hardly worth building all those lanes just for that blip of time.
Better to spend a fraction
of that money making them safe in all lanes.

Besides, special lanes, at-grade, don't really work. If you don't
trust the vehicle on
ordinary road, how is it going to cross intersections and crosswalks?
>

That's why I looked at the DARPA urban challenge results and said,
"now it looks
like operation on city streets is doable, if we put the will into
it." If you don't accept
that, then you can keep talking about PRT, but you have to answer the
question,
why is it that 40 years later there's no PRT, if it's really so good?

The prediction that we can build these safe vehicles is of course not
a sure one.
But it seems to be reasonably credible, and thus I am examining what
it means,
if and when it happens.

But anything that happens from the bottom up will quickly surpass
anything done from
the top down, like special lanes, PRT, rails or other transit. Of
that I am pretty confident.

And of course, there is my intermediate "whistlecar" vision, which
does not require
vehicles safe enough to drive people, just vehicles capable of moving
themselves, more
slowly, for refueling, parking and self-delivery.
> Jay

[Jerry Roane]

Brad

I am sorry if my sand comment bent your feathers.  I can see from your web presence you were promoting one of your web essays.  Just highlight the sand sentence hit the del key and continue on.  I am interested in your ideas.  If solar version of TriTrack is fiction then robo cars are just as much fiction hey?  The fact that steel wheels running on guideway combined with cars with a wind tunnel tested Cd of .07 allows solar power generated and stored in the battery mules will power the entire transportation grid.  It is only the extreme energy efficiency of guideway that allows this to be.  Your stating "crazy" and "silly" just indicates that you have not run the numbers yet.  If you want to learn I am sure you will like the consumer level calculator on our web site.  It lets you input various Cds, frontal areas and weights on street cars and steel wheeled cars and provides mathematical answers for energy expenditure.  Using Sanyo panels at 17% energy conversion efficiency provides 100% of the electrical power needed to move 2.8 million people in Denver using Denver sunshine data averaged over the year under the cars but above the crowd.  As you astutely point out that same power could come from the power grid but by combining solar panel power with extremely energy efficient guideway cars the cost is disguised and as you must know getting the quantity of solar panels built up brings down the unit price.  Unless someone moves first we are doomed to the biggest wealth transfer in the history of wealth to butcher a slogan. 

The percentage of energy we waste of the sunshine is easier to explain to the public if we are solar-only.  Tesla is still working on their motor/controller efficiency.  If you want hard data go to UQM for their solar challenge motors and controllers.  They built many of the motors that went across Australia at 40 mph using solar power.  The statement that transportation is silly that is solar powered is disproved by the solar challenge race series.  How do you gel this with the assertion?  If instead of placing a million dollar solar panel on a few square feet of the roof of the car you put 127 feet by 7 feet of panel under the guideway that is instantly dedicated to that one car then you can see that solar can power that one car faster than the solar challenge cars using UQM motors. 

This is not a debate contest so you can put your feathers back down.  This is an important discussion with our future in the balance.  I am not against computer controls.  That is in my field.  My opinion is not to overstate what controls can do in the general public till you have a working tested system.  There are several on this list doing exactly that and some of them will succeed.  When they have their testing done then robo things can be shown in a working state.  If you want to control a system it is wise to reduce the control complexity and degrees of freedom.  As you know from software "bugs" AKA coding mistakes are a function of the programmer's skill and the complexity of the task.  If the task is unnecessarily complex then you need a better programmer and better programmer has the upper limit of available humans trained in the art.  If Bill Gates can't find a programmer who can write an email program then how can you find a programmer to entrust the very lives of millions of people driving home?  Guideway provides a path to robo cars if that is the ultimate goal. 

Your energy assertions are currently close but we intend to rock that world soon.  Stay tuned.

Jerry Roane
CEO Roane Inventions Inc.

snip

Whoops.  That sort of talk ends any debate with me.  If you would care
to repost your
thoughts without this sort of statement in them, and pledge never to
make such statements

again in your apology, I will read them.  I hereby pledge never to ask which sand your head is placed into. 

[Dennis Manning]

Brad:

You wrote:

<If you don't accept that, then you can keep talking about PRT, but you have
to answer the question, why is it that 40 years later there's <no PRT, if
it's really so good?

If I'm not mistaken the idea for a robocar goes back to the 30's. I think
it's you that should answer the question. Besides PRT is here or coming
soon: Morgantown, ULTra underconstruction, Vectus nearing test completion,
and Masdar looking promising.

One more question. How fast do you envision robocars? Have you ever checked
out Applied Levitation? or et3?

Dennis

[Brad Templeton]

> Are you suggesting robocars require no drivers license and every one can
> afford a robocar thereby eliminating public transit? Surely you jest.
>
> BTW if a computer is controlling the car who is liable in an accident?
>

No jest at all, have you read the main robocar articles? Seen the
videos?

I forecast a very low price for robotaxi rides, with a base operating
cost under
10 cents a mile and falling. (Under 1 cent/mile for electrical
energy, and a
depreciation cost highly dependent on the cost of the vehicles and
their lifetime and
maintenance costs.) It seems reasonable that this could be sold
profitably at
30 cents/mile, which would be -- without any special subsidy other
than the existence of roads -- cheaper than the typical transit ticket
for most trips.

So yes, I think it can be very affordable.

There is a large section on the question of liability.

[Brad Templeton]

On Sep 6, 4:28 pm, "Jerry Roane" <jerry.ro...@gmail.com> wrote:
> Brad
>
> I am sorry if my sand comment bent your feathers.  I can see from your web
> presence you were promoting one of your web essays.  Just highlight the sand
> sentence hit the del key and continue on.  I am interested in your ideas.
> If solar version of TriTrack is fiction then robo cars are just as much
> fiction hey?

Solar isn't fiction, it just does not make sense to talk about solar
powered transportation. Solar is a source of electricity, like
coal, NG, nuclear, hydro, wind etc. What you are talking about (and
so am I) is electric powered transportation, which is indeed good.

One could, in theory, set up a transportation system (or a house, or
a building, or a factory) that is powered by solar and not connected
to the grid, and call it "solar powered." But that would be silly, if
your goal is to be green. The proper thing to do with solar is to
grid-tie it, so that every kwh generated by the solar is put to use
(and indeed, put to use at the peak time.) If you don't grid-tie
your solar, you get times when the power is just being discarded
because your system is not using it, and that's anti-green.

Since few who install solar want to be anti-green, it would not make
sense to build solar powered transportation. You can build electric
transportation (good) and you can generate renewable energy for the
grid (good) where you will reduce the demand for coal (really good)
but these are orthogonal.

Off-grid solar makes sense only where there is no grid at all, which
is not usually the case in urban transport design. Off-grid solar in
a city is, I will reiterate, worse than silly.

Or do you speak of some kind of solar-thermal driven car? Where
thermal energy drives the wheels? I had not heard of anything like
that before.

> went across Australia at 40 mph using solar power.  The statement that
> transportation is silly that is solar powered is disproved by the solar
> challenge race series.  How do you gel this with the assertion?

These are contest vehicles. A solar panel on a moving car will,
again, often find itself throwing away its output because the car is
not moving and the batteries are not highly discharged. That's
what's silly -- building expensive solar panels and then discarding
the power they produce in the name of "being pure solar." If you
need a rural vehicle that can't recharge, or a contest vehicle which
will always run when there is sun, then solar can make sense, but
that's not what we are discussing here -- which is urban transport.
In a city you can readily charge batteries from the grid, or draw
power from a guideway if you have that.

 If instead
> of placing a million dollar solar panel on a few square feet of the roof of
> the car you put 127 feet by 7 feet of panel under the guideway that is
> instantly dedicated to that one car then you can see that solar can power
> that one car faster than the solar challenge cars using UQM motors.
>

> available humans trained in the art.  If Bill Gates can't find a programmer
> who can write an email program then how can you find a programmer to entrust
> the very lives of millions of people driving home?  Guideway provides a path
> to robo cars if that is the ultimate goal.

This would make sense -- if anybody had shown success at deploying a
large guideway infrastructure. I wrote a paper on PRT for my urban
geography class in 1976. I must say I'm tired of waiting, and I'm
sure I am not alone. Watching how innovation works in the computer
industry, I have become more and more of the opinion that a chance for
transportation to follow that path is more likely to bring success.

I've seen many explanations about why there are no real PRTs (and just
a couple of toy PRTs at Heathrow and Dubai) but they have the smell,
to me, of "there's always an explanation, but never a success."

I believe the answer is that cities -- which are who buys transit --
are not innovators. It is the exception rather than the rule. Even
the toy PRTs at Heathrow and the UAE being built not by cities, but by
an airport with a larger budget, and a slightly crazy but visionary
dictator. Morgantown is not PRT as I see it.

If this is correct -- that it is the resistance to innovation among
municipal transit agencies that blocks the adoption of PRT -- then it
has a long hill to climb. Cities won't buy it until they see a large
working system. But a large working system isn't anything like what
we see. A large system has 100s of km of guideway, and 100s of
stations, because it needs to take a lot of people from where they are
to where they want to go, not be like traditional transit and take
them only between a few selected popular places.

[Brad Templeton]

On Sep 6, 6:17 pm, "Dennis Manning" <john.manni...@comcast.net> wrote:
> Brad:
>
> You wrote:
>
> <If you don't accept that, then you can keep talking about PRT, but you have
> to answer the question, why is it that 40 years later there's <no PRT, if
> it's really so good?
>
> If I'm not mistaken the idea for a robocar goes back to the 30's. I think
> it's you that should answer the question. Besides PRT is here or coming
> soon: Morgantown, ULTra underconstruction, Vectus nearing test completion,
> and Masdar looking promising.

Until last year, people thought of robocars as pure science fiction.
Then DARPA held their contests and we got to be astounded by what
could be done for a prize that is tiny in the world of transportation
($1M and $2M.) And DARPA (and the war) demonstrated a need for early,
expensive tech for moving cargo in a war zone. It is these contests
that started people talking seriously, after watching real robocars in
action, about what they might do and mean.

40 years from now, if they are still 10 years away, we can make the
same complaint.

You really call Morgantown with its single line, 5 stations and giant
cars PRT? I call it a people mover.

What I thought people meant by PRT was a system with many lines, many
stations, and point to point non-stop transfer between stations for
individuals or small groups.

Is that not what PRT means?

My current suspicion is this. If somebody can get a PRT system
underway, it won't be up and running as a full sized, many station PRT
for some time -- a decade at least.

And around the time it is, I think we will start seeing robocars ready
to take to ordinary streets. And people will say, "Can we tear down
those annoying PRT lines?" Or at least adapt them for the 3.5' wide
robocars as extra lanes. But if they are blocking the sky, I think
people will ask to tear them down.

It was this timeline that led me to conclude PRT may become obsolete
before it gets to exist. As soon as you attribute a significant
probability that robocars will deploy within about 12-15 years, you
should not build a PRT (unless you hope to make it be robocar track in
the future.)

>
> One more question. How fast do you envision robocars? Have you ever checked
> out Applied Levitation? or et3?

For urban use, probably 35mph. On highways, same as today's cars.
For fast intercity travel, I expect people to continue to use planes,
or trains where they are appropriate.
>
> Dennis

[Dennis Manning]

Brad:

Thanks for responding. I guess we will just have to watch the development of
PRT/DM/Robocar. You really should see et3 to see what's possible for PRT.

Goodluck.

Dennis

----- Original Message -----
From: "Brad Templeton" <bra...@gmail.com>
To: "transport-innovators" <transport-...@googlegroups.com>

Sent: Saturday, September 06, 2008 7:12 PM
Subject: [t-i] Re: PRT obsoleted by robocars

[Jerry Roane]

Brad

I am glad to see you engage.  I am sure when you think through the solar concept you will see the showmanship and the value of the idea.  It is showmanship to propose solar power for transportation but I do need to correct some of your assumptions.  The main assumption is that we waste energy.  We do not.  The massive number of batteries needed to energy sip thoughout the system will store all the incoming energy.  On days when there is more energy input that output the battery mules will feed the power grid during peak energy use.  This means instead of selling electric power at below wholesale to the monopoly like present solar installations we sell it to the grid at peak daily price.  Since this is a network of cars with history we know how much power to sell to the power grid on any one day.  Because this is a large amount of energy we will be getting market rates not a discounted version.  You are correct that solar kilowatts are more expensive this year than grid power but as pollution credits and carbon offsets come along our system will be immune from these "sin taxes".  This is a global concept that pits the US against the rest of the world and we will be out voted.  The air pollution aspect of a grid powered transportation system makes it muddy in the mind of the public.  By going full solar there is no misunderstanding that the power to run the cars is from only sunlight.  Again this is only possible to be solar only if the cars are very energy efficient.  I saw where you were saying under 1 cent per mile energy cost.  Those are some light weight cars with slippery tires and a small frontal area.  Do you have such a car built?  Has it been tested in a wind tunnel for total drag.  I can run the numbers for you if you have the basic vehicle parameters.  At some slow speed a gasoline car can get 1000 mpg but most people want faster cars than the supermileage competition.  

"

 If you don't grid-tie
your solar, you get times when the power is just being discarded
because your system is not using it, and that's anti-green."

I think this choice of expression is invalid.  If sunshine impinges on the Earth and no one harvests that energy is that anti-green?   The incorrect assumption is how much battery storage we intend to provide via the millions of battery mules under the guideway end points.

"

Off-grid solar in
a city is, I will reiterate, worse than silly."

What is silly is being afraid to convert to solar when the technology is here.  We just need to apply it.  Germany has taken the lead in the field and they are wisely buying up all the world solar panel production.  For the next 25 years these panels will be producing wealth for their nation while we atrophy.

"Or do you speak of some kind of solar-thermal driven car?   Where
thermal energy drives the wheels?  I had not heard of anything like

that before.'

Electric power is the middle transition method.   Once energy is in electric form it is easy to move it to other forms like vehicle motion.  The thing to come may be more of what you are asking about but my legal team is not finished with the latest.  When it gets its date stamp it can be shared.  You will like it.

> went across Australia at 40 mph using solar power.  The statement that
> transportation is silly that is solar powered is disproved by the solar
> challenge race series.  How do you gel this with the assertion?

These are contest vehicles.  A solar panel on a moving car will,
again, often find itself throwing away its output because the car is
not moving and the batteries are not highly discharged.   That's
what's silly -- building expensive solar panels and then discarding

read ahead rather than ask just before the answer. 

This would make sense -- if anybody had shown success at deploying a
large guideway infrastructure.  

Que the choir!  That is what we propose.  I am not a Pure PRT type because pure PRT is using the hardware only one way.  We will let the market determine how the hardware is used.  We just provide the efficient hardware.  Our third patent explains how to cheaply deploy guideway using an automated system of guideway manufacture.  The funding is by individuals mostly which is in line with your thinking about how a new system can come to be financially.  The city only has to give permission. The guideway is so inexpensive that it can be paid for with a light mileage fee in a short time.  The key to large scale deployment is super low cost in my view.  These partents are newer than 1976 but they have their origins in the first oil embargo. 

"

I wrote a paper on PRT for my urban
geography class in 1976.  I must say I'm tired of waiting, and I'm
sure I am not alone.   Watching how innovation works in the computer
industry, I have become more and more of the opinion that a chance for
transportation to follow that path is more likely to bring success."

I agree fully.  Once transportation goes computer age we will see miraculous things.  

I've seen many explanations about why there are no real PRTs (and just
a couple of toy PRTs at Heathrow and Dubai) but they have the smell,
to me, of "there's always an explanation, but never a success."

I believe the answer is that cities -- which are who buys transit --
are not innovators.   It is the exception rather than the rule.   Even
the toy PRTs at Heathrow and the UAE being built not by cities, but by
an airport with a larger budget, and a slightly crazy but visionary
dictator.  Morgantown is not PRT as I see it.

If this is correct -- that it is the resistance to innovation among
municipal transit agencies that blocks the adoption of PRT -- then it
has a long hill to climb.  

We are working this with gusto.  We are making some headway believe it or not.  Just by being part of the wall paper eventually they will hear the message.  Understanding will be the next step of course.

 

Cities won't buy it until they see a large
working system.  But a large working system isn't anything like what
we see.  A large system has 100s of km of guideway, and 100s of
stations,

Ours does not have stations only merge zones. 
 

because it needs to take a lot of people from where they are
to where they want to go, not be like traditional transit and take
them only between a few selected popular places.

I agree fully on this point too.  Transportation can be more ideal than it is with the help of automation.

Jerry Roane 

[Ian Ford]

Brad Templeton wrote:
> Solar isn't fiction, it just does not make sense to talk about solar
> powered transportation. Solar is a source of electricity, like

> coal, NG, nuclear, hydro, wind etc. [snip]

I think everyone here gets that a mix of sources powers the grid, and
electric transport is and should be grid-tied. However, some believe
that "solar transport" is a political selling point, even if it involves
exploiting some incomplete understanding of elected officials of how
electricity works.

I agree with your point that cybercars can more easily jump the
introduction threshold than any new fixed rail mode can.

Along those lines, I've been thinking in the past year or so that the
most feasible leap in technology is a combination of BRT and
neighborhood cybercars. BRT with long station spacing and priority
signals is a practically instant, cheap and effective way to serve the
distance in metro transit, but it lacks any solution for the last mile.
Cybercars going 15 mph, which do NOT use arterials but only neighborhood
streets and urban core streets, would cost effectively serve the last
mile and could be put in service fairly soon it seems. I estimate that
these two companion technologies could move US cities from 1-5% transit
share to 10-25%, and lower overall transport costs.

From that situation, two competing trends could occur: (1) cybercars'
speed limits could be relaxed and they could use arterials, thereby
providing end-to-end transport; and (2) the BRT lines could be upgraded
to a routable, sometimes elevated, fixed rail network for higher speeds
and less congestion. Both of those trends could occur and find an
equilibrium along with the existing modes; I don't see why one of them
has to necessarily win out over the other.

The step after THAT would be to upgrade cybercars to be able to use the
routable network. This sequence of trends almost eliminates the
introduction thresholds.

[gary]

Brad,

> As soon as you attribute a significant
> probability that robocars will deploy within about 12-15 years, you
> should not build a PRT (unless you hope to make it be robocar track in
> the future.)

So in 15 years robocars will be traveling at highway speed?
Coexisting with human controlled vehicles, no less? I think my
approach of running at grade, on rails, and WITHOUT automobiles
(greatly simplifying the task) is possible in that timeframe, but I
think you're way too optimistic (assuming I understood you correctly).

On a related note, have you spoken with any of the DARPA vehicle
developers? Do they share your accessment of the technology?

gary
www.PRTProject.com

[Kirston Henderson]

on 9/6/08 11:49 PM, Ian Ford at i...@ianford.com wrote:

> Along those lines, I've been thinking in the past year or so that the
> most feasible leap in technology is a combination of BRT and
> neighborhood cybercars. BRT with long station spacing and priority
> signals is a practically instant, cheap and effective way to serve the
> distance in metro transit, but it lacks any solution for the last mile.
> Cybercars going 15 mph, which do NOT use arterials but only neighborhood
> streets and urban core streets, would cost effectively serve the last
> mile and could be put in service fairly soon it seems. I estimate that
> these two companion technologies could move US cities from 1-5% transit
> share to 10-25%, and lower overall transport costs.

Ian,

I hate to have to get into the sales mode for our systems, but all of
this fantasy dreams of robocars discussion makes it impossible for me to
refrain because I know how to accomplish most of the goals right now, so
please forgive me and it any of you are offended, just delete this post and
don't read further.

The MicroRail system of which we first road tested the first dualmode
vehicle on Labor Day can nicely solve that last mile problem with a
combination of core, main lines carrying a combination of passenger trains
and CarTrains carrying compact automobiles that drive onto and off of the
CarTrain CarFerries at special stations along the main lines. If the small
cars are built with compatible automatic electrical connectors on the
underside, they can be low-speed, pure electric vehicles that nicely solve
that last mile problem. Furthermore, this technology is here now and
initial systems could start operating within about three years. Because the
main lines and trains carry both passengers, cargo, and small personal
automobiles, such systems should earn enough revenue to recover their
initial cost and pay all of the O&M costs, something that few conventional
mass transit systems have ever done. Furthermore, we don't have to wait for
any more technology development such as robocar technology, that may or not
ever become a practical reality. The guideway is low enough in cost, that
most cities could easily purchase and install these systems with revenue
bonds and probably never spend a dime of tax money.

By the way, we have already found ourselves giving two different Fort
Worth city council members our presentation and shown them the vehicle and
they came to us without even before we were able to start any marketing
effort, an activity that we just started this last week. I believe that
this indicates a real interest in some public transportation officials in
lightweight, elevated systems that do far more than conventional systems and
do it at far less cost.

With such a system in place, I suspect that public transit share of
travellers within cities might rather rapidly grow to more than 50% and in
the process, greatly reduce the haze of heavily carbon product polluted air
in large cities. It would also greatly decrease the trip times for most
users. Use of such a system would also greatly decrease the travel cost for
most users. With the combination of 95% efficiency motors and regenerative
braking from the motors, we were amazed that after about 30 minutes of
driving the car in the pure electric mode, the could not even detect any
drop in the battery voltage by the end of the driving test.

I suggest that you all take a careful look at the following url:

http://megarail.com/pdf/MCPCAL-3.pdf

By the way the above url may change within the next few days. However,
the only change will be from a -3 to a -4.

Kirston Henderson
MegaRail®

[Brad Templeton]

> correct some of your assumptions.  The main assumption is that we waste
> energy.  We do not.  The massive number of batteries needed to energy sip
> thoughout the system will store all the incoming energy.  On days when there
> is more energy input that output the battery mules will feed the power grid
> during peak energy use.

So that's a grid tie. I am not sure why the batteries though. Many
people seek to allow the storage of excess off-peak power to use in
the grid during peak, and generally batteries are not considered a
workable method. I would be curious what numbers you have that would
say otherwise. If not, why would you have batteries? It would be
disappointing to see people using batteries to store solar in a grid-
tie system just for the sake of saying they are solar, since batteries
have their own environmental costs, though they are not particularly
great, they are not zero.

Right now solar is not grid-competitive in most places (it is at high
tiers in California) without the rebates/tax breaks, and it's not yet
competitive with the rebates/breaks in many places, though this is
changing. When it become competitive, I expect to see a lot more
solar.

However, this is a distraction, showmanship as you say. What we're
talking about is electric driven transportation, it doesn't matter
where it comes from.

> sunlight.  Again this is only possible to be solar only if the cars are very
> energy efficient.  I saw where you were saying under 1 cent per mile energy
> cost.  Those are some light weight cars with slippery tires and a small
> frontal area.  Do you have such a car built?  Has it been tested in a wind
> tunnel for total drag.  I can run the numbers for you if you have the basic
> vehicle parameters.  At some slow speed a gasoline car can get 1000 mpg but
> most people want faster cars than the supermileage competition.

Electric cars are able to do better than 1 cent/mile today. Indeed
electric
trikes can go over 30 miles on a kwh, which costs an average of 9
cents in the USA.

> I think this choice of expression is invalid.  If sunshine impinges on the
> Earth and no one harvests that energy is that anti-green?   The incorrect
> assumption is how much battery storage we intend to provide via the millions
> of battery mules under the guideway end points.

Because solar panels take energy to manufacture (about 1-2 years of
their output) the worst off-grid solar systems (ie for camping) use
more energy than they generate. This would not be the case for
systems like this. What's anti-green would be not doing grid-tie when
it's available, because any power you toss away rather than putting
into the grid means less coal is burned. It is anti-green to burn
more coal just so you can say you're solar.

However, you are doing grid tie based on your descriptions, you just
have batteries, presumably because of pricing regulations.

>

> What is silly is being afraid to convert to solar when the technology is
> here.  We just need to apply it.  Germany has taken the lead in the field
> and they are wisely buying up all the world solar panel production.  For the
> next 25 years these panels will be producing wealth for their nation while
> we atrophy.

Germany pays something like 40 cents/kwh. Solar beats that, even at
today's prices.

[Brad Templeton]

On Sep 6, 10:20 pm, gary <garydst...@gmail.com> wrote:
> Brad,
>
> > As soon as you attribute a significant
> > probability that robocars will deploy within about 12-15 years, you
> > should not build a PRT (unless you hope to make it be robocar track in
> > the future.)
>
> So in 15 years robocars will be traveling at highway speed?

Sure. In 1995 a german experimental robocar traveled the autobahns of
Germany with human oversight, and occasional human intervention.
However the computers drove the car, not simply at highway speed, but
autobahn speed, 93% of the time. 12 years ago. The Darpa grand
challenge 2 cars drove desert back roads at not quite highway speeds,
I think they could easily do it at that speed now. Highway driving is
actually the easy part. It's urban driving that will be the real
challenge.

> Coexisting with human controlled vehicles, no less?

Of course. Pointless if they can't do that. In the Urban challenge
they did do that, but not at highway speeds.

 I think my
> approach of running at grade, on rails, and WITHOUT automobiles
> (greatly simplifying the task) is possible in that timeframe, but I
> think you're way too optimistic (assuming I understood you correctly).

We won't be rid of human driven automobiles at grade for a few
decades, at least one after the robocars are introduced.

>
> On a related note, have you spoken with any of the DARPA vehicle
> developers?  Do they share your accessment of the technology?

Yes, some of them see 2020 as a target date.

[Brad Templeton]

>
>     I hate to have to get into the sales mode for our systems, but all of
> this fantasy dreams of robocars discussion makes it impossible for me to
> refrain because I know how to accomplish most of the goals right now, so
> please forgive me and it any of you are offended, just delete this post and
> don't read further.
>

This is one of my key points. There is what we can design or build,
and there's what will actually get bought. In particular there is
what cities will buy, and it's a tiny subset of what we can design,
dream or even prove.

This is why I predict robocars as the eventual winner. Because they
will not be bought by cities, but by people. Initially rich
technophiles with money to burn. They buy stuff in a pattern so
different from how cities buy things it's night and day.

It's been possible to build basic PRTs for many years now -- for
decades if you believe PRT advocates. Is the failure of cities to
buy them simply because the PRT advocates were wrong and the
technology is only now becoming possible? Or is there something in
the way cities by transit that makes PRT hard to sell? I think at
this point they feel like the citizens of Springfield being sold a
monorail.

[rob...@2getthere.eu]

I think the idea of robocars is very realistic. Especially considering
the fact that you see more and more Advanced Driver Assistance systems
being installed in cars. These developments include a couple that every
driver has gotten accustomed to: your navigation system, parking sensors
and now the sensor that automatically parks the car based on sensors.
Surely you'll see more of these systems introduced in the high-end
models first (e.g. the heads-up display) and trickling down to other
models eventually...

At the same time PRT is obviously gaining interest and ground. Basically
it's public transit becoming more private, like the car, in an automated
version. Seems to me these are two converging development lines,
ultimately ending at approximately the same point (depending on the
technology applied for the PRT systems...)

But I also believe they will have to co-exist. PRT will be the public
transit system, while robocars will the private system. The advantage
PRT has over robocars is significant in one aspect: parking. Robocars
would still require dedicated parking spaces everywhere, wasting
valuable space, while PRT vehicles would be re-used. That's why they
could co-exist, as do cars and public transit today. It would be great
if both systems could share the same infrastructure, not requiring
dedicated lanes for either one.

If you like PRT couldn't robocars be the next thing? If you like
robocars couldn't be PRT systems be the step-up to it? You can disagree
about the time-line (when does what get done), but this will not be
limited by technology, but rather by emotions (of users, customers and
decision makers...)

Robbert

gary schreef:

[gary]

Brad

> Yes, some of them see 2020 as a target date.

Can you elaborate on what they had to say? I've had a couple of brief
email exchanges with a couple of the engineers, but didn't get
anything so concrete as this.

gary

[eph]

Brad,
Back in 2000, there was a "heavy truck platooning" project which seems
to have been successfully demonstrated. The idea was to have a series
(2 or more) trucks wirelessly connected and driven solely by a lead
driver - think of the potential savings! Somehow, 8 years later,
nothing has come of it.
http://ivsource.net/archivep/2000/aug/a000831_chauffeur.html
I'd like to know what you think of this.

F.

[Bruce Attah]

Brad, I have several quibbles with your robocar idea:

First, I disagree with the implied statement in the following
question: "So how to you compare a technology that's not here yet,

with a technology that's seemed possible for 40 years but never
realized?"

Second, I disagree with the idea that "PRT's few remaining advantages

(better efficiency through electrified rails and less congestion on
custom-built ROW) barely blip the radar."

Third, I disagree with the claim that "PRT, like all transit, has to

be bought by risk-averse municipal transit planners with decade long
time horizons and huge budgets."

Regarding the first point: PRT has seemed possible for 40 years, and
was first realized in a working scale model 36 years ago (Aerospace
Corporation), and in full, 30 years ago (Cabinentaxi). That first
realization was not commercially adopted for reasons that boil down to
the fact that it was expensive. The Cabinentaxi PRT system worked
perfectly, passed all its tests, including endurance tests, and
received the thumbs up from the regulatory authorities. However, the
system was designed to support both PRT and APM, which means it was
bulky and complicated, and would not be able to undercut rival APM
systems. The second realization (Raytheon, about 13 years ago) also
worked, technically, but also was too expensive, this time because of
some decisions that made no engineering sense, which resulted in
vehicles and guideway being extremely bulky (6000 kg for a 4-passenger
vehicle - six times the weight of a Smart ForFour, and 2.5 times the
weight of a Bentley Continental).

So, the technical viability of PRT has been demonstrated. Only its
economic viability has, until recently, not been. This is quite a
different situation from the robocar, whose technical viability is
still a far-off dream. Moore's law does not apply to AI, by the way.
All major AI projects (natural language processing, robotic mimickry
of human motion, computer vision, etc.) are currently decades behind
the expectations of pioneer researchers. The rule with AI is that any
problem involving real-time pattern matching in a real-world setting
is at least one order of magnitude more difficult than you think it is
when you start out. I don't think we can expect the robocar, in the
sense of an autonomous vehicle that can reliably and safely mix with
all traffic in all conditions will be with us in much less than twenty
years time. The DARPA challenge merely confirms this. The fastest
vehicle travelled at an average 15 mph, on uncongested roads with no
pedestrians, and with the help of 3000 waypoints alongside to aid
guidance. There's a gulf between that and riding on ordinary roads in
ordinary urban traffic. The same applies to the tests done in the
1990s in Germany of a computer driven car that drove on the autobahn,
and was fine for quite long distances in that setting (no cross-
traffic, no pedestrians or cyclists, no sudden junctions, human driver
on hand to take over if anything unexpected occurs). It's a long way
from that to true autonomy.


Recently, the economic viability of PRT has been demonstrated. ULTra
in the UK, and Vectus in Sweden, have built test tracks (both now
certified safe for carrying passengers), and shown that this can be
done cheaply. Their vehicles and guideways are of sensible size. ULTra
has sold its first system to a private sector customer (within 3 years
of launching its test track - so no decade-long time horizon, there),
and far more cheaply than an equivalent APM would cost (about 1/3rd of
the price). PRT is now cheap enough for both municipal and private
sector clients to consider buying. What's more, independent studies
indicate that its life-cycle cost is lower than that of equivalent
train, bus, light-rail and tram systems.

I know that, according to your website, the ULTra system is not PRT.
You do not explain why you think so, but according to the commonly
accepted definition of PRT, the ULTra system conforms 100%. I suppose
you must be using your own personal definition. ULTra is also, in a
limited sense, a robocar. It is able to drive and steer itself on a
flat surface, using lasers and sonar for guidance.

Regarding the second claim, that electrified rails and custom ROW are
a trivial advantage, I think you are making a massive error.
Electrified rails offer infinite range (no stopping for fuel or
battery charge, no matter how long the journey), and offer total
energy flexibility (no need to worry about where the fuel is coming
from, or whether the oil is running out). Exclusive guideways mean
never having to deal with cross-traffic, or drunk, inattentive,
incompetent, or sleepy human drivers. Robocars mixing with human-
driven traffic and pedestrians would still get stuck in traffic, and
would still get caught in accidents through the fault of others. PRT
would be immune to these problems.

Regarding the third claim, that only municipal clients can buy PRT,
the facts show you're wrong. The first client for PRT so far has been
a private sector client, and there are at least hundreds, perhaps
thousands, of other potential private sector clients for similar
systems. PRT is feasible at small scales where APMs, light rail, etc.,
cannot compete. The existence of such a client base create the
opportunity for competition and innovation that you claim is only
available to robocars.

Finally, in the long run there is no difference between robocars and
PRT. If robocars come to dominate, they will eventually push human-
driven traffic off the roads (the latter will come to be considered
too dangerous). The roads will get narrower, to take advantage of the
robocars' ability to drive straight and steady, and to reduce road
building and maintenance cost. The robocars will sprout power pickups,
and so gain the twin advantages of infinite range and engergy-source
agnosticism. Among the first customers for robocars will be taxi
companies and car hire firms. Due to the elimination of drivers, and
reduced insurance costs, it will become as cheap to ride a taxi as to
ride a bus. In consequence, individuals will cease buying private
cars, because ownership will confer few or no advantages over simply
hiring a vehicle on a per-use basis, and will actually have some
distinct disadvantages (e.g., why wait for your own car to return from
picking up the kids, when you can jump into a robotaxi that is already
nearby?). End result: computer-driven vehicles on exclusive guideways,
run by transport service providers, offering point-to-point, non-stop,
non-scheduled service -- i.e., PRT.

[Bruce Attah]

Two more points:

> And people will say, "Can we tear down those annoying PRT lines?" Or at
> least adapt them for the 3.5' wide robocars as extra lanes. But if they are
> blocking the sky, I think people will ask to tear them down.

I think the opposite is more likely: roads at ground level are a pain.
People hate walking alongside them, and don't particularly delight in
crossing them. big roads divide cities up, separating former
neighbours.

Meanwhile, PRT guideways do not "block the sky". The ULTra guideway is
1.5 m wide and 45 cm deep (60 inches by 18 inches). The Taxi 2000
guideway is 42 in by 42 in (approx 1m by 1m), and the Vectus guideway
is based on a 20 inch pipe. These would cast very little shadow, be
barely noticeable from a distance, and would in any case be easy to
hide or disguise. The Mist-er guideway, with its open structure, would
cast even less shadow than the others. The supports required for all
these systems are about 50cm (20 in) or less wide.

Elevated PRT is therefore a totally different proposition from
elevated highways or light rail (anything from 40 ft wide upwards, and
several feet thick, with enormous supports, and casting huge gloom
wherever they go).

> Even the toy PRTs at Heathrow and the UAE being built not by cities, but by
> an airport with a larger budget, and a slightly crazy but visionary dictator.

I'm not sure what is supposed to be crazy about the several sheiks and
sultans behind the Masdar project, and I do not presume to
psychologize them from afar. The project itself is very sane, given
that UAE needs to diversify its economy *before* the world shifts away
from oil dependency. Perhaps you know why you call these people
slightly crazy.

What I don't understand is why you insist on calling the Heathrow and
Masdar projects "toys". In the case of Masdar, you're talking about a
PRT network that is supposed to be the main transport system for a
city of 50,000 people. How is that a "toy"? In the case of Heathrow,
you're talking about a system that replaces a very expensive shuttle-
bus service at one of the busiest airports in the world, and is
planned, after pilot, to extend to encompass the surrounding area
(hotels, business parks, etc.), with 400 vehicles. This project is as
big as the entire public transport system for many medium sized towns.
Airport administrators have an abiding hatred of shuttle buses,
because of their high operating costs and poor quality of service. A
successful pilot in Heathrow opens up a market to hundreds, possibly
thousands, of applications for PRT in the very near future.

[Bruce Attah]

I did not completely spell out what I wanted to say about elevated
versus at-grade: roads, being a pain, and elevated PRT being only
slightly visually intrusive, but not being intrusive at all in other
ways, means that people may prefer elevated over ground level
transport. Many roads at ground level could be torn up, and replaced
by naturally-draining grass and flower beds, and walkways and
cycleways from which motorized vehicles are completely excluded. This
potential improvement in the urban space means that instead of people
saying "let's tear up the annoying PRT and build ground-level roads",
we will more likely find them saying, "let's tear up the roads and
build elevated PRT".

[Walter Brewer]

Like electronic warfare, the measures-countermeasure game goes on and on.
How about Roboshield for pedestrians/bikers? And maybe some animals
Like aircraft collision avoidance systems, warnings, and maybe shutdowns
Robocars when paths are predicted to cross.

And a Robobumper; airbag like that catches and protects the victim?

Walt Brewer

----- Original Message -----
From: "Bruce Attah" <bruce...@googlemail.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Sunday, September 07, 2008 11:28 AM
Subject: [t-i] Re: PRT obsoleted by robocars

>

[Jerry Schneider]

At 09:49 PM 9/6/2008, you wrote:
>snip -----------------------------------

>Along those lines, I've been thinking in the past year or so that the
>most feasible leap in technology is a combination of BRT and
>neighborhood cybercars. BRT with long station spacing and priority
>signals is a practically instant, cheap and effective way to serve the
>distance in metro transit, but it lacks any solution for the last mile.
>Cybercars going 15 mph, which do NOT use arterials but only neighborhood
>streets and urban core streets, would cost effectively serve the last
>mile and could be put in service fairly soon it seems. I estimate that
>these two companion technologies could move US cities from 1-5% transit
>share to 10-25%, and lower overall transport costs.

This sounds like the French effort to do the same thing with the
radial RER commuter rail
lines in the Paris region. The project involved developing little
cars that could be used to
access the RER stations. To get the cars back to their origins, they
would be led by
a driven vehicle and follow it in a platoon, linked electronically, I think.
There is lots of info about Cybercars at www.cybercars.org
Also: http://www.cybercars.org/cyb-technologies.html#platoon
The EU people appear to be far ahead of the flashy DARPA effort in terms of
their thinking, problem solving and development/demonstration efforts.

- Jerry Schneider -
Innovative Transportation Technologies
http://faculty.washington.edu/jbs/itrans

[rob...@2getthere.eu]

Reading Bruce's post I figured there should be a way to put the
discussions we keep having about elevated vs. at grade to rest for
once-and-for-all:
- elevated guideways are more dominant in the land- and city scape than
guideways at grade
- great care is taken to minimize the amount of visual intrusion by
minimizing the guideway
- privacy issues are being addressed to prevent objections of residents
- elevated guideways do provide the opportunity to re-model the land-
and city scape at grade
- elevated guideways are likely to be required in densily populated
areas (no space at grade)
- elevated guideways are less likely to be embraced by the public in
residential area's (emotion)
- elevated guideways should not be introduced into historic sites (but
provide access nearby)

Is the above agreeable to all? There are contradictions, but those will
always remain (you have to take them into account when designing the
system and trying to convince decision makers). Any additions?

Robbert

Bruce Attah schreef:

[Brad Templeton]

I've seen the BOSS team (which won the latest challenge) give that
date to the press. Sebastian Thrun has also given reasonably near
term dates though I don't know if I've heard him say a particular
year. It is foolish to name a specific year -- there are so many
variables that can push it back (or, due to advances in other fields
of AI, bring it forward) and a lot of the variables are political
rather than technical in my view -- as in when will people be
ready. For example, if they are made safe enough so that they kill
500 vs human drivers killing 45,000 woudl society take that
numerically good deal? Quite probably not. What number they would
actually take is hard to predict.

[Brad Templeton]

That is one of the things on my roadmap page.

On Sep 7, 8:27 am, eph <rhapsodi...@yahoo.com> wrote:
> Brad,
> Back in 2000, there was a "heavy truck platooning" project which seems
> to have been successfully demonstrated.  The idea was to have a series
> (2 or more) trucks wirelessly connected and driven solely by a lead
> driver - think of the potential savings!  Somehow, 8 years later,

> nothing has come of it.http://ivsource.net/archivep/2000/aug/a000831_chauffeur.html

[eph]

Kirston,
That is very interesting - allowing the "trains" to run as regular
road vehicles, DM trains if you will.

It seems implied that PRT cybercars could utilize MicroRail instead of
the larger MegaRail system?

Does the efficiency of the rail vehicles suffer if snow tires are
needed instead of low rolling resistance tires? How do you keep
potholes and such from causing damage to the vehicles/tires? If the
trains run on battery power when off the rails, how far can they go?

I can see great advantage to having all wheels drive (esp. in snow/ice
conditions). Traction-control might also be useful. There are
problems with articulated buses in Ottawa. Worst example:
http://farm3.static.flickr.com/2260/2301963552_c62163152e_m.jpg
http://busdriverofdurham.blogspot.com/2008_03_01_archive.html

F.

On Sep 7, 2:44 am, Kirston Henderson <kirston.hender...@megarail.com>
wrote:

[eph]

Yes, the point was more that after 8 years of technical availability,
there has been zero adoption. The next chapter on school of fish was
fantastic. If a tree hits the first car, a string of them will
crash. That's the brickwall stop discussion we frequently have.
Complete avoidance of obstacles seems particularly outlandish.
Computers have only recently managed to best chess grandmasters at a
fairly well understood game. Let's hope the systems won't be running
a Microsoft Operating System when someone steps in front of a
vehicle. I think it will take a long time for people to accept
robocars without some physical barriers and backups. Not saying it
won't happen, but there is time to fully exploit a PRT/DM system in
the interim. In fact, I think the PRT/DM systems will be the gateway
to fully automatic vehicles for that reason.

F.

[Brad Templeton]

On Sep 7, 8:28 am, Bruce Attah <bruce.at...@googlemail.com> wrote:

>
> Regarding the first point: PRT has seemed possible for 40 years, and
> was first realized in a working scale model 36 years ago

We are using a different meaning of realized. For me, it's all about
successful commercial adoption. Many technologies are possible, but
only some actually win in the market and become the transportation
means of millions. Since one of my prime goals is to find a way to
more efficient transportation, only something that can become the
transport for a billion people is of interest, and that requires
marketing success, not technological demonstration.

If we had a city with a large PRT network, and lots of the population
using it, people deciding they don't need a car because the PRT is
there, that would be a realization of the PRT vision. The PRT vision
was never, in my view, taking people to their airline terminal or
running a single line with a few stops like a light rail. Many
people here seem to be thinking of PRT as a technical problem. While
it is not without technical challenges, it is really a marketing and
deployment problem.

,

> still a far-off dream. Moore's law does not apply to AI, by the way.
> All major AI projects (natural language processing, robotic mimickry
> of human motion, computer vision, etc.) are currently decades behind
> the expectations of pioneer researchers.

Yes and no. Moore's law, which is a marketing law rather than a
technological one, applies to only some of the driving components
behind AI, namely raw CPU and parallel processing. But you might be
surprised to learn that there's been much more algorithmic improvement
than you think, especially in the part of AI needed for robocars,
which is machine vision. Some algorithms have improved so much that
today's code on 20 year old computers would outperform the 20 year old
code on today's computers. The forecasting errors of the pioneers
just tell us they had not grasped the problems at that time. We
haven't yet either, but we're a little better at it.

> years time. The DARPA challenge merely confirms this. The fastest
> vehicle travelled at an average 15 mph, on uncongested roads with no
> pedestrians, and with the help of 3000 waypoints alongside to aid
> guidance.

I fully expect robocars to have complete waypoint roadmaps, why would
they not?
Doing it faster is one of the problems Moore's law does help with.
But yes, these were prototypes, but built for tiny budgets and there
is far to go. It could be that we hit a wall as we try to do
better. But what if we don't hit too many such walls? What does
that mean for transportation?

>

>
> I know that, according to your website, the ULTra system is not PRT.
> You do not explain why you think so, but according to the commonly
> accepted definition of PRT, the ULTra system conforms 100%. I suppose
> you must be using your own personal definition. ULTra is also, in a
> limited sense, a robocar. It is able to drive and steer itself on a
> flat surface, using lasers and sonar for guidance.

It needs a special ROW, it can't go on an ordinary street. When I
talk of robocars, I mean ones that don't need any modification of the
roads.

It has a whopping THREE stops on a single 4km line. 18 pods. Top
speed 40 km/h. That's not a real PRT though it does for the first
time implement the personal part.

To me the PRT vision was always an attempt to give people many of the
good things of private vehicles together with the good things of
transit. That requires a system, not a line, since one of the good
things about the car you want is that it goes from A to B, you don't
wait to change lines.

>
> Regarding the second claim, that electrified rails and custom ROW are
> a trivial advantage, I think you are making a massive error.
> Electrified rails offer infinite range

They offer infinite operating time, but of course are limited to the
area covered by the rails. When we talk of infinite range, I mean
"distance does not affect whether it can take you there."

Electric transportation is good, of course. Batteries have their
costs, but their go-anywhere benefits are huge, and the public really,
really likes them. Electrified lines only take you where electrified
lines go, and of course they cost lots of money to build.

I list electrified lines as a minor advantage because the cost of
batteries while not zero, is judged as well worth the freedom they
provide. If somebody can develop a design so that private vehicles
could safely tap into electric lines along the street it would be
great. Even cities are using hybrid diesel-electric buses for their
transit lines because putting in dedicated electric lines is only
worth it in the densest areas.

Dedicated ROW is great of course, but it comes at an even larger cost
than the lines. Dedicated ROW is also wasteful unless it gets used
as much as shared roads, once you have systems on the shared roads
that manage congestion better.

My robocar vision, by the way, has no stopping for refuel/recharge
because you say your destination in advance. You thus use a vehicle
which has the capacity to take you there. To the user, it has the
appearance of arbitrary range. This means if they say they are going
40 miles into the burbs, they probably get a liquid fuel vehicle, not
an electric one, depending on how good battery tech is. But that's a
minor portion of trips.

> would still get caught in accidents through the fault of others. PRT
> would be immune to these problems.

Indeed. Private ROW is great, who wouldn't want it? But at what
cost?

There is one other downside to private ROW, which is that it's almost
always single lane per direction, which means that stalled vehicles,
obstacles, suicides etc. shut down the entire line. On roads one can
just drive around obstructions.

> Regarding the third claim, that only municipal clients can buy PRT,
> the facts show you're wrong. The first client for PRT so far has been
> a private sector client

Do you really want to hold up Heathrow as an example of the PRT
vision? It just doesn't fit my view of it. The private systems will
largely tend to be like that. A small number of lines (perhaps only
one) and a small number of stations used by a captive audience (where
there is no competing means of transportation.)

A real transit system has different problems. It needs to serve lots
of people in a large area, and must compete with their other means of
getting around (cars, taxis, carpools, existing transit lines)

Yes, it would compete with the existing transit lines run by the same
transit authority. For I see some of them saying, "Well, the dowtown
core is already well served by the light rail, so let's have the PRT
just feed the light rail."

>
> Finally, in the long run there is no difference between robocars and
> PRT.

Yes, as I say in the essay on PRT, you can view robocars as an
implementation of the PRT vision. But there is one huge difference
between robocar PRT and the PRT most people talk about. It's who
buys it and owns it. And that turns out to be the largest difference
of all, because it makes a huge difference in the pace of innovation
and how well it really serves the desires of the customers. It's the
top-down central planning vs. bottom-up market-driven difference. It
dwarfs all the other differences and so you can't say there is none.

(There are a few other ones, such as going anywhere, even driveways
and dirt roads vs. only going approved place etc.)

Among the first customers for robocars will be taxi
> companies and car hire firms.

They will be good customers, but no, the first customers -- and this
is supremely important -- will be rich early adopters.

In fact, robocar technologies are already being sold and planned to go
into conventional cars to make them accident resistent. These show up
first in a Lexus or BMW, 5 years later they are in a VW or Toyota.

Rich early adopters are what made the tech revolution what it is.

Due to the elimination of drivers, and
> reduced insurance costs, it will become as cheap to ride a taxi as to
> ride a bus. In consequence, individuals will cease buying private
> cars, because ownership will confer few or no advantages over simply
> hiring a vehicle on a per-use basis, and will actually have some
> distinct disadvantages (e.g., why wait for your own car to return from
> picking up the kids, when you can jump into a robotaxi that is already
> nearby?). End result: computer-driven vehicles on exclusive guideways,
> run by transport service providers, offering point-to-point, non-stop,
> non-scheduled service -- i.e., PRT.

I don't think people will cease to buy their own vehicles. First, I
predict robocars will become cheaper than today's cars. Secondly,
since you can hire out your robocar to others if you like, you can
afford a better one. Ahyway, I go into lots of detail on this in the
web site.

But it's not PRT as people think about it, while it does fill the
role. The "T" right now stands for transit, a word that typically
means public transportation in common usage.
However, it does implement the vision, because it provides all the
advantages of transit and all the advantages of private cars in one
package. It's why I am advocating it.

As for blocking the sky: There have been much good work in making
elevated guideways that are not too obtrusive, but you can't argue
they don't have a visual impact. Hell, overhead power, cable and
telephone lines have a bad visual impact, no matter how neat you make
them, and they are just wires, they don't block the sky. And no way I
want a guideway looking into my bedroom on the 2nd story where before
it was only my neighbour 3x as far away.

And yes, the Masdar system, if it gets built, is not so much a toy, I
was thinking of the Dubai one. And no, the Sheik isn't really
crazy, he's one of the world's mavericks. My point is it takes
somebody like that to sell a PRT today.

[Brad Templeton]

That's one of the things that makes prediction in computer/electronics
tech so complex. When you double your capacity every year, the first
year you go 1 mile and the 10th year you hit 1000 miles, but you did
500 of those miles in the final year. So we can really only look at
the broad trends well, not the specifics.

The school of fish (or even swam of bugs, you don't need very advanced
intelligence to do this) do not all crash when an obstacle appears.
That's the point. You can dive through the fish and not touch a one
no matter how hard you try. The fish have evolved patterns of
behaviour to always leave the gaps needed to get through this.

Safe robocar operation will always have lots of gaps in the traffic
flow (as does safe human driving operation!) and those will always
been monitored, and there will be conversations with the other cars
about them. HDVs will have larger gaps near them. Sidewalks will
have a gap of course. You design it so that if anything appears into
the flow of traffic (stopped car, pedestrian, dog, etc.) that you can
always make use of a gap to not hit it. If somebody hides behind
something and suddenly jumps out to suicide, you may not be able to
avoid it, but I think you can do a very good job on most things,
better than humans.

(PRT on guideways depend on the guideway but suffer much worse if
something or sombody suddenly appears on the guideway.)

With enough gap, you can even avoid the clever suicide, possibly with
some jarring to your passengers from the sharp turn, especially if the
other vehicles around you are listening to your warnings and judge you
trustable. It's a question of how much gap we want vs. how much we
want to pack a road to its fullest capacity. Humans can only go so
fast on foot, faster on bikes, faster yet in cars. I think switching
to one-way streets makes a lot of sense as it solves the oncoming
traffic problem better.

[eph]

On Sep 7, 3:42 pm, Brad Templeton <brad...@gmail.com> wrote:
> That's one of the things that makes prediction in computer/electronics
> tech so complex. When you double your capacity every year, the first
> year you go 1 mile and the 10th year you hit 1000 miles, but you did
> 500 of those miles in the final year. So we can really only look at
> the broad trends well, not the specifics.

Given that stopping distance increases with the square of speed, does
this mean the exponential growth factors cancel out? If this is the
case, a simple linear projection of time (vehicle doubling speed
requires a simple time factor times computer doubles speed). I don't
really know the specifics... Is my math wrong? ;-)

>
> The school of fish (or even swam of bugs, you don't need very advanced
> intelligence to do this) do not all crash when an obstacle appears.
> That's the point. You can dive through the fish and not touch a one
> no matter how hard you try. The fish have evolved patterns of
> behaviour to always leave the gaps needed to get through this.

But the fish aren't constrained by 2 dimensional, lane width
restricted path. I think Computers will eventually exceed humans in
this task, but zero crash seems optimistic. I would bring that back
to "best option chosen" every time, as defined by the rules imparted
to the system.

>
> Safe robocar operation will always have lots of gaps in the traffic
> flow (as does safe human driving operation!) and those will always
> been monitored, and there will be conversations with the other cars
> about them. HDVs will have larger gaps near them. Sidewalks will
> have a gap of course. You design it so that if anything appears into
> the flow of traffic (stopped car, pedestrian, dog, etc.) that you can
> always make use of a gap to not hit it. If somebody hides behind
> something and suddenly jumps out to suicide, you may not be able to
> avoid it, but I think you can do a very good job on most things,
> better than humans.

Sure. I agree, eventually. On the other hand, if PRT/DM guideways
are available and (high-speed) travel becomes significantly safer than
HDV, the bar will be higher for robocars. Also, guideway travel has
potential to raise the bar by operating at higher speeds than our
current highways.

I certainly see low speed cybercars roaming residential areas unmanned
in a short time. In fact, it's part of how I see PRT being used.
They will probably have big yellow striped lanes where pedestrians
must beware at first (shifting the onus) and eventually be accepted as
"normal" traffic.

>
> (PRT on guideways depend on the guideway but suffer much worse if
> something or sombody suddenly appears on the guideway.)

Who says guideways won;t be monitored? The same technology used in
cars can be used to keep the guideway safe. The guideway itself can
advise the vehicles of obstacles instead of (or complementing) on-
board sensors. This could reduce costs and increase reliability
further.

>
> With enough gap, you can even avoid the clever suicide, possibly with
> some jarring to your passengers from the sharp turn, especially if the
> other vehicles around you are listening to your warnings and judge you
> trustable. It's a question of how much gap we want vs. how much we
> want to pack a road to its fullest capacity. Humans can only go so
> fast on foot, faster on bikes, faster yet in cars. I think switching
> to one-way streets makes a lot of sense as it solves the oncoming
> traffic problem better.

It's a trade-off to be sure. G-forces the occupants must/can endure
are also a consideration. Hopefully there won't be as many suicides
in our future.

One-way streets have some virtues, but not really needed or desirable
for residential/low traffic areas.

F.

[Marsden Burger]

Bruce et al:

 

First, I mean no disrespect to anyone on this board, I enjoy monitoring the dialog.  At the same time, after monitoring the development of transit technology since 1969 (actually that is also the way I see this board) I have a bit of understanding regarding this field.

 

Which brings me to a point:

 

"There is nothing more effective than an undeveloped transit system!"

 

If I did not create this phrase, I have surely used it quite a bit, and the truth of it has never diminished. 

 

(Actually, I think maybe George Pastor did.  He was good at coining phrases - "People Mover" was one that he started that he always wished he hadn't, but how do you pull it back after it caught on?)

 

I bring up this phrase, because it is very appropriate for all of those here that are doing their best, through transportation innovation, to find ways to improve our society and make the quality of life better for citizens around the world - and that is the ultimate goal that ties us all together. 

 

Before a system goes into operation carrying thousands of people a day, there is a long road ahead - hats off to those of you who have advanced from the talking stage to the hardware stage. 

 

I jump into these discussions now because of Bruce's statements:

 

 

    Regarding the first point: PRT has seemed possible for 40 years, and
    was first realized in a working scale model 36 years ago (Aerospace
    Corporation), and in full, 30 years ago (Cabinentaxi). That first
    realization was not commercially adopted for reasons that boil down to
    the fact that it was expensive. The Cabinentaxi PRT system worked
    perfectly, passed all its tests, including endurance tests, and
    received the thumbs up from the regulatory authorities. However, the
    system was designed to support both PRT and APM, which means it was
    bulky and complicated, and would not be able to undercut rival APM
    systems.

 

First I have every hope that we will find applications for the Cabintaxi technology and get it and further updated versions into operation.

 

Going back to history for a second.

 

    "was not commercially adopted for reasons that boil down to
    the fact that it was expensive"

 

    In the little writing I have done, I have stated clearly the reasons that Cabintaxi was not built in Hamburg.  Money always is a factor in everything, but the Cabintaxi system costs were not the reason for the Cabintaxi system not being installed.  The superficial reason was the program costs not falling in line with the funding available in the budget of the Ministry for Research and Technology.  That budget was severely impacted by an across the board 10% cut in all federal departments of German government at the time.  When transportation development efforts switch from test facilities to real world implementation, the budget demands for a program skyrocket.  Where as Cabintaxi was only one of a hundred programs under the umbrella of the Ministry of Research and Technology during it normal development effort, the real world implementation made the Cabintaxi project 10% of the total program of the Ministry, and by cutting that one project - only one program to deal with negatively - met the 10% program reduction requirement of the government directive.  Is this the reason - yes.  Is this the entire reason in detail, of course not - it is just true that it was not the costs per mile of the Cabintaxi system that caused it not to be installed.

 

      "The Cabinentaxi PRT system worked
    perfectly"

 

    Not even I would claim this!

 

 

    "passed all its tests, including endurance tests, and
    received the thumbs up from the regulatory authorities"

 

    This is essentially correct.  As good as engineering gets, engineering will always be the art of correcting mistakes.  At the point of approval for the installation in Hamburg, Cabintaxi was at the end of a 10 year development program - six iterations of full scale built and tested guideway cross sections and 14 iterations of full scale vehicles; designed, developed, built, run and tested, refined, and redesigned.   We were going into the Hamburg project with a seventh iteration of the guideway, and we would alter the vehicle undercarriage on the upper running system to stiffen the undercarriage's relationship to the vehicle control.  (I make this point to again demonstrate that refinements will always be needed, and that thinking that a system can go from a simple test track to an actual installation, has historically been wishful.)

 

     "which means it was
    bulky and complicated"

 

    Ahmmm.....tell me a transit system that is more complicated than PRT, adding 12 passenger or 18 passenger vehicles which can operate in a PRT mode does not make the system more complicated than PRT, actually they allow the system to be implemented in less complicated programs to allow a network system to get started....show me a developed PRT system that is less bulky than a single lane of Cabintaxi........show me any system that can give bi-directional urban mass transit operation at the cost of a Cabintaxi bi-directional system....show me a single level urban area PRT network that can not be installed more effectively and at lower cost with an over-and-under Cabintaxi PRT network...show me anyone that has truly studied what the Cabintaxi system actually represents......

 

    "and would not be able to undercut rival APM
    systems."

 

    Just so we are clear on how I can make the following statements, I led the Mannesmann and Messerschmidt joint venture marketing efforts for the Detroit People Mover Project, and after Cabintaxi withdrew from the transit field, I went to work for UTDC and was the first corporate representative in Detroit to start the implementation of what became the present Detroit system.  While the joint venture backed away from placing a bid on the Detroit project, we had already worked out an in house bid.  That bid was twenty percent lower than what the winning bid was from the APM competition.  While that bid was twenty percent lower, it still provided the bi-directional operation - not requested but desired by Detroit.  Our bid (company policy) was to have been a fixed price bid (even though that was not part of Detroit's RFP) with our joint venture taking the risks of over-run, and building it into our price, still twenty percent less than the competition.  This created a situation where we would have been proposing a system with twice the guideway miles, more vehicles providing superior service levels, and still doing it at twenty percent less than the competing APM suppliers running a single lane system.  That does not even take into consideration the Detroit DPM project overran - surprise surprise - by 100%. 

 

To the present:

 

No PRT system made it as far in its implementation as the Morgantown system. Yet the rush to its implementation means it will never be built again.  Endurance testing on the back of the customer resulted, in Morgantown, in a program that was able to get a system running, but not one that the developers would ever build again.  Cabintaxi may have been on the other extreme, it might have benefited from a quicker installation program so that the funding agency did not experience administration changes where those newly involved had little commitment.  

 

There are many things in human nature and transit development that are not directly obvious, years spent in this business lets me know that debating is of little use.  I just did not want these statements, understandably not meant in negative intent, to be accepted as fact.

 

I am not aware of any urban PRT system, in concept or under development, that at the time that development is finished, will represent a more effective urban transit system than the already developed Cabintaxi technology.  If we are successful in our ongoing efforts, our intent is to privately operate advanced transportation systems.  If and when we are successful, show us a better system, and we would look to include it in our operational efforts.  Private e-mails are always welcome.

 

Sincere best wishes,

 

Marsden Burger

Cabintaxi Corporation   

 

 

--- Original Message -----

From: Bruce Attah

To: transport-innovators

Sent: Sunday, September 07, 2008 11:28 AM

Subject: [t-i] Re: PRT obsoleted by robocars

[Jack Slade]

Do you also have the answer to finding the half - trillion dollars that is needed to fix the roads "right now", money that nobody has because they have spent it on other projects. Then you may need a similiar amount over the next 40 years to repair them again.   This may require a trip back to the dreaming board.   Jack Slade --- On Sat, 9/6/08, Brad Templeton <bra...@gmail.com> wrote: From: Brad Templeton <bra...@gmail.com> Subject: [t-i] PRT obsoleted by robocars To: "transport-innovators" <transport-...@googlegroups.com> Received: Saturday, September 6, 2008, 8:43 PM In response to some discussion on this group of my robocar articles in relation to PRT, I wrote an expanded section on why I make the rather bold claim that a not-yet practical technology like robocars could make us declare a never-really-implemented technology like PRT to be on the verge of becoming obsolete. It's at http://www.templetons.com/brad/robocars/prt.html I started out as a fan of PRTs basic ideas -- personal vehicles, little waiting and non-stop station to station transportation which would give riders something much closer to what they get from cars, while keeping many of the benefits of transit (no driving, no parking, limited congestion.) I even dreamed of hybrid PRTs that left the track and were driven the last mile on city streets. But then, thanks to DARPA, robocars started moving from SF to reality and could offer more than PRT ever promised without requiring any new infrastructure. So how to you compare a technology that's not here yet, with a technology that's seemed possible for 40 years but never realized? The answer for me came by looking at who pays for it. Robocars can be privately owned and funded, with very minimal government involvement. As their abilities improve with Moore's law, they will be bought by rich early adopters, then bought again next year by the same early adopters. This approach drives incredible, unbeatable innovation. PRT, like all transit, has to be bought by risk-averse municipal transit planners with decade long time horizons and huge budgets. It is almost a rule that what they buy will be obsolete by the time they break ground, let alone by the time it's in operation. Anything that brings Moore's law, individual purchase, early adopters and highly competitive markets to transportation is sure to win. I don't think there's even a contest. PRT's few remaining advantages (better efficiency through electrified rails and less congestion on custom-built ROW) barely blip the radar. Thus the bold statement. Give it a read for more details.

[Brad Templeton]

On Sep 7, 4:34 pm, Jack Slade <skytrek_...@rogers.com> wrote:
> Do you also have the answer to finding the half - trillion dollars that is needed to fix the roads "right now", money that nobody has because they have spent it on other projects. Then you may need a similiar amount over the next 40 years to repair them again.
>  
> This may require a trip back to the dreaming board.

People want roads, and even in the most optimistic PRT or transit
projection, people will want and demand roads. Funding will be found
for what people scream for. I have good confidence in this.

Now it turns out there are some interesting elements of robocars that
might allow them to tolerate roads that are worse, such as complete
detailed maps of all potholes and other road flaws (to the cm)
combined with adaptive suspension which is thus aware of upcoming
bumps and, if it can't steer around them, can make them not felt by
the car.

This is because while an HDV needs a tight "feel the road" suspension,
a robocar wants a mushy "hide the road" suspension, or at least the
passengers do. There is interesting work going on in ferromagetic and
other adaptive suspension technologies that could produce a smooth
ride, so long as the driver doesn't need to feel the road.
However, that's a sidebar. I am confident we will have drivable roads
for decades to come. Robocars, fully deployed, would save several
trillion dollars in other ways, so finding the money would be possible.

[Bruce Attah]

Marsden,

Thanks for the gen on Cabintaxi and the reasons for its non-adoption
by the intended clients. It is intriguing that the APM project won
against Cabintaxi in Detroit even after making a higher bid. And the
DPM is a loop, so so I would expect the space-saving Cabintaxi design
to be at an advantage there. Did Morgantown queer the pitch? Or was it
not relevant?

Meanwhile, I think it might be fair to say that a PRT guideway that
has vehicles attached both above and below is a bit more complicated
than one that supports vehicles only from above, or only from below,
and an APM system that also offers PRT is more complicated than a pure
APM. How much this would affect costs in actual implementations, I
don't know. I can see the possibility that a Cabintaxi system of two-
way loops could be cheaper in a high-capacity application than a rival
PRT system that used one-way guideways. In another application, where
a simple one-way guideway provided adequate capacity, I imagine that
the Cabintaxi system would tend to be somewhat more expensive (perhaps
only slightly). Is this fair, or am I wildly off base?

PS, by "perfectly", I of course did not mean "unimprovably" -- merely
that it did what it said on the tin.

[Brad Templeton]

On Sep 7, 2:09 pm, eph <rhapsodi...@yahoo.com> wrote:
> On Sep 7, 3:42 pm, Brad  Templeton <brad...@gmail.com> wrote:
>

> But the fish aren't constrained by 2 dimensional, lane width
> restricted path.  I think Computers will eventually exceed humans in
> this task, but zero crash seems optimistic.  I would bring that back
> to "best option chosen" every time, as defined by the rules imparted
> to the system.

Correct, but we can be even smarter than bugs or fish, I would
venture.

Consider a road with just a few cars. You would agree that there is
enough space in such a road that the cars, with lightning reaction
times, could swerve, break and accelerate as needed to avoid any human
trying to get near them. The question is, with good algorithms, how
many cars can you add to the road before that is not true? And once
you add that many, if there is not enough capacity, how much risk are
you willing to take to add more capacity? It's not a question of can
it be done, I think, but rather of how much capacity it costs to leave
the gaps to keep it so safe. It can certainly be that safe during
off-peak, the only question is rush hour. There are many solutions
which offer themselves at rush hour, however, including of course
having a little transit left over.

>
> Sure.  I agree, eventually.  On the other hand, if PRT/DM guideways
> are available and (high-speed) travel becomes significantly safer than
> HDV, the bar will be higher for robocars.  Also, guideway travel has
> potential to raise the bar by operating at higher speeds than our
> current highways.

I don't see there being a bar. If the robcars are taking you nonstop
door to door, the alternative would have to be very much faster to
even get the customer's attention. In such events, we would see a
system where people take robocars to high speed train stations, get on
the high speed trains and then get into waiting robocars at the other
end. This would only apply to people moving from suburbs, since
over shorter urban distances, the transit could be a teleporter and
not be worth the transfers compared to a short trip.

For intercity, I do expect high speed transit and planes to be the
norm. Planes become much nicer with robocars, since a security
screened robocar is waiting for you on the tarmac when your plane
lands -- no terminal needed at all, just a weather cover.

>
> I certainly see low speed cybercars roaming residential areas unmanned
> in a short time.  In fact, it's part of how I see PRT being used.
> They will probably have big yellow striped lanes where pedestrians
> must beware at first (shifting the onus) and eventually be accepted as
> "normal" traffic.

I am not sure we will be ready to shift the onus onto the
pedestrians. The public is too afraid of computers. The public will
want to know their baby can crawl into the street and be safe from the
robots. And since babies are slow, they will be, early on.

> Who says guideways won;t be monitored?  The same technology used in

Of course. But once you have your sensors and machine vision this
good, who needs guideways? I get the sense that one of the reasons
for guideways, grade separated ones, is that people don't trust having
people and vehicles that close to one another, it's a substitute for
sensors.

>
> It's a trade-off to be sure.  G-forces the occupants must/can endure
> are also a consideration.  Hopefully there won't be as many suicides
> in our future.

Suicide by transit is one of the more popular methods, alas.

[Bruce Attah]

Brad Templeton wrote:
> On Sep 7, 8:28�am, Bruce Attah <bruce.at...@googlemail.com> wrote:

> It has a whopping THREE stops on a single 4km line. 18 pods. Top
> speed 40 km/h. That's not a real PRT though it does for the first
> time implement the personal part.

The first PRT application has to start as a pilot. For anyone without
bottomless pockets, it would be a mad gamble to build 50 km of
guideway with a thousand vehicles before operational testing and
refinement of a small pilot had been completed. Pilot doesn't mean
toy, though. It's a real application, with considerable expansion
planned for later phases.

The point is, it's a start. That puts PRT some distance ahead of
robocars in the race to implementation.

> To me the PRT vision was always an attempt to give people many of the
> good things of private vehicles together with the good things of
> transit. That requires a system, not a line, since one of the good
> things about the car you want is that it goes from A to B, you don't
> wait to change lines.

I think there is every prospect that such systems will exist within
about five years of now. All that is required is a successful pilot.
As soon as this happens, there will be a veritable flood of
implementations.

> They offer infinite operating time, but of course are limited to the
> area covered by the rails. When we talk of infinite range, I mean
> "distance does not affect whether it can take you there."

Yes, and electric rails offer inifinite range. There's no road that's
longer than the Trans-Siberian railway, which is electric.

Not that electric rails are part of the definition of PRT, mind.

> Yes, as I say in the essay on PRT, you can view robocars as an
> implementation of the PRT vision. But there is one huge difference
> between robocar PRT and the PRT most people talk about. It's who
> buys it and owns it. And that turns out to be the largest difference
> of all, because it makes a huge difference in the pace of innovation
> and how well it really serves the desires of the customers. It's the
> top-down central planning vs. bottom-up market-driven difference. It
> dwarfs all the other differences and so you can't say there is none.

I do say there is none, because as I said, private ownership of cars
will almost totally disappear, if robocars become the reality.

> (There are a few other ones, such as going anywhere, even driveways
> and dirt roads vs. only going approved place etc.)

0.1% of trips. No consequence.

> Among the first customers for robocars will be taxi
> > companies and car hire firms.
>
> They will be good customers, but no, the first customers -- and this
> is supremely important -- will be rich early adopters.

You are not contradicting what I said. I said that *among* the first
customers will be the taxi companies and car hire firms.

> I don't think people will cease to buy their own vehicles.

I think they will, for the reasons I've given.

> First, I predict robocars will become cheaper than today's cars.

Why?

> Secondly, since you can hire out your robocar to others if you like, you can
> afford a better one.

That is a *very* unattractive prospect, and I don't think it will
happen. Suppose there's 90% car ownership in my town. What's the
prospect that if I leave my car to pick up fares while I'm in the
office, I will earn more than chump change from doing so? Not great.
What's the prospect, meanwhile, that the passengers picked up will
leave their litter in my car, or dirty it with muddy feet, or that
they will riffle through any documents I happen to have left
carelessly in the car, Pretty good. In my private car? No thanks. The
whole sense of the car being private is tainted, that way. Either my
private car is genuinely private, or I may as well hire a cab, and not
bother with ownership.

> As for blocking the sky: There have been much good work in making
> elevated guideways that are not too obtrusive, but you can't argue
> they don't have a visual impact.

I don't claim that they're invisible, I claim that they're not ugly
(or certainly don't have to be, at least). They're not intrusive in
that negative sense.

> Hell, overhead power, cable and
> telephone lines have a bad visual impact, no matter how neat you make
> them, and they are just wires, they don't block the sky.

They're ugly, nothing to do with blocking the sky.

> And no way I want a guideway looking into my bedroom on the 2nd story where before it was only my neighbour 3x as far away.

Singapore, where all the trains are elevated, solved this problem some
years ago. When one of their trains goes close to a residential
building, the windows, which are of switchable glass, become opaque.

> And yes, the Masdar system, if it gets built, is not so much a toy, I
> was thinking of the Dubai one. And no, the Sheik isn't really
> crazy, he's one of the world's mavericks. My point is it takes
> somebody like that to sell a PRT today.

Yes, it takes a maverick to pioneer something fairly radical, but it
doesn't take a maverick to be the second or third client to adopt, if
the first purchase worked out well, or at least was prestigious. The
pioneering spirit is not strong in municipal authorities, but the
copycat spirit is very strong indeed.

[eph]

One of the promises of robocars is platooning to reduce freeway space
needed, so that number of vehicles vs safety trade-off is
predetermined. Highways cost more than a few extra fatalities
(sadly).

You want people to transfer from a robocar to a public vehicle then
transfer back to a different robocar? Compared to getting into your
car and driving all the way to your destination, I'd call that a step
back. I propose low speed electric vehicles with access to guideway
bogies for high-speed travel. No switching cars. Ownership is
possible, though not necessary. Low speed PRT or cybercars are
possible.

Exclusive guideways aren't a substitute for sensors, they're insurance
against snafus. Guideways would also allow operation in adverse
weather conditions. I think it will take some time for people to
build trust in the new systems. Operation within guideways builds
that confidence and insures that those growing pains won't be fatal.
Remember how well air-bags worked at first? Child sitting in the
front seat, air-bag deploys - oops. People have reason to fear new
(untested by experience/time) technology.

F.

[Kirston Henderson]

on 9/7/08 1:54 PM, eph at rhaps...@yahoo.com wrote:

> Kirston,
> That is very interesting - allowing the "trains" to run as regular
> road vehicles, DM trains if you will.
>
> It seems implied that PRT cybercars could utilize MicroRail instead of
> the larger MegaRail system?

Ans: The MicroRail guideway will be able to carry dualmode vehicles
specially designed for operation on the guideway. It may not matter as to
how they are controlled during street operations.

>
> Does the efficiency of the rail vehicles suffer if snow tires are
> needed instead of low rolling resistance tires? How do you keep
> potholes and such from causing damage to the vehicles/tires? If the
> trains run on battery power when off the rails, how far can they go?

Ans: I would expect that dualmode vehicles equipped with snow treads
would work well on the guideway, but would likely generate a little more
noise. As for pothole damage, we expect that dualmode vehicles will be
limited to reasonably low speeds and thus be less susceptible to damage from
potholes. The dualmode trams would all be equipped with on-board generators
plus batteries. Small dualmode personal vehicles might well be battery
powered and designed for ranges of only about 15 miles without charging or
returning to the rail.

>
> I can see great advantage to having all wheels drive (esp. in snow/ice
> conditions). Traction-control might also be useful.

Ans: Our vehicles are driven by two or more electric wheelmotors and
include traction control in the control systems.

> There are problems with articulated buses in Ottawa. Worst example:
> http://farm3.static.flickr.com/2260/2301963552_c62163152e_m.jpg
> http://busdriverofdurham.blogspot.com/2008_03_01_archive.html

Interesting stuff.

Kirston Henderson
MegaRail®

[eph]

As a safety/resilience feature, the ability to leave the guideway
allows, for example, the ability to circumvent a disabled section of
guideway.

Another possibility which I know you don't endorse would be an at-
grade system with crossings. The ability to leave the guideway to
cross an intersection (for vehicle crossings) would allow MegaRail to
be installed practically anywhere LRT might be considered. It could
have rail-like crossing barriers to increase it's exclusivity.
Probably such an at-grade system would be much lower speed of course.

I suppose the generator would likely be smaller than a regular hybrid
bus' because lower speeds would be perfectly acceptable.

F.

On Sep 7, 10:26 pm, Kirston Henderson <kirston.hender...@megarail.com>
wrote:

[gary]

This has been a fascinating dialog! I thank Brad for kicking it off.
In looking for info on the DARPA technologies, I found a VERY
interesting video. It's a full 40 minutes and starts out slow, but I
highly recommend it to all of you...

http://blip.tv/file/920488

As I now understand traditional PRT and Brad's RoboCar theory, I find
my system (www.PRTProject.com) right smack in the middle. I'm
proposing running at grade, but still on rails for numerous reasons
already discussed. The above video may be a real eye opener for some
of you. What I took away from it was that the technology my system
required was almost certainly LESS than ten years out. Even
approaching off-the-shelf!

gary

[Brad Templeton]

> Yes, and electric rails offer inifinite range. There's no road that's
> longer than the Trans-Siberian railway, which is electric.

I'm not sure what you are getting at. People really care "Can this
vehicle take me where I am going?" If they are going to a station on
the trans-siberian, then I am sure it is an excellent choice. They
care can this vehicle take me on this trip. That it can keep going
for somebody else is not relevant.

>

> I do say there is none, because as I said, private ownership of cars
> will almost totally disappear, if robocars become the reality.
>
> > (There are a few other ones, such as going anywhere, even driveways
> > and dirt roads vs. only going approved place etc.)
>
> 0.1% of trips. No consequence.

Alas, not by today's buying patterns. Today people buy a car to meet
all their needs (that they can afford.) They buy an SUV if they go to
the hills twice a year. They buy a pickup if they haul cargo a few
times a year. They won't buy a car with 100 miles range if they go
120 miles once a month.

I have predicted a drop in car ownership as the need for it is
relaxed, but I think it's much too far to predict the complete end of
private car ownership. If cars are affordable and are not a burden to
store, there will be car ownership for some portion of society that
wishes it, and for another portion they will always hire cars. I am
confident of this prediction for quite some time to come.

> > First, I predict robocars will become cheaper than today's cars.
>
> Why?

First of all, 1 and 2 person vehicles become much more marketable, and
they actually are what you need for most trips. Secondly, since you
don't need long range in most vehicles (as most trips are short) the
battery needs are far more modest. Without an expensive battery,
electric cars are simpler and cheaper than gasoline cars, and will get
even cheaper with mass production. Finally, once car crashes are a
thing of the past (which is further in the future) you can build
vehicles without all the anti-crash systems (crumple zones, roll bars,
air bags etc.) making them much cheaper. A basic 1-person 3-wheeled
electric car with fiberglass shell, suitable only for non-highway
driving, can probably be built for well under $2,000. The computer
part follows Moore's law and is eventually just a few hundred dollars,
though it costs much more to start.

> prospect that if I leave my car to pick up fares while I'm in the
> office, I will earn more than chump change from doing so? Not great.

That's why the market will balance ownership vs. hiring. You were
saying most people will want to hire -- that makes it good prospects
to own.

Your car will largely have a per-mile operating/depreciation cost if
well used. If your commute is 20 miles/day, and you hire it out for
an additional 20 miles per day, but at double your per-mile cost, it
would in fact make a profit for you.

If it is possible to have a profitable robotaxi business with full
time robotaxis, it is obviously possible to have those vehicles be
owned by private owners who allocate them first to their own use and
hire them out when not used. This is particularly true for people who
don't have a traditional commute and thus can hire out their vehicle
at the peak demand time when the spot price is highest.

> What's the prospect, meanwhile, that the passengers picked up will
> leave their litter in my car, or dirty it with muddy feet, or that
> they will riffle through any documents I happen to have left
> carelessly in the car, Pretty good. In my private car?

Have you read my essays?

> > Hell, overhead power, cable and
> > telephone lines have a bad visual impact, no matter how neat you make
> > them, and they are just wires, they don't block the sky.
>
> They're ugly, nothing to do with blocking the sky.

They are there in my view, and that's something people don't like.
They also will prefer getting into and out of vehicles at ground level
over having to ascend or descend, though of course in multi-storing
buildings with integrated PRT station this is not a problem.

[Dennis Manning]

Maybe to summarize a bit on the threads of ground level vs elevated and robo
vs PRT.

The common ground is that we are all talking automation - removing the
driver in favor of the computer. We seem to diverge on the question of the
best path/guideway, and the questions about what can best solve the
implementation challenge.

Further I think there's a primary question. If you have a vehicle that is
computer controlled then what is the best path/guideway. I'd simply submit
that it isn't a slab of pavement.

Brad says it is. Gary says it's rail at ground level, and many say it's
elevated guideway.

Maybe this ferment is what keeps us on our toes. I'd just add that all of us
are serving up better options than what we are getting from our respective
governments, or the corporate private sector for that matter.

What we all need is a winning demo of an automated network. It has to be a
network. Elevators are automated. That's not what we are all about. What
it's all about is an automated network.

One last comment. It's not about about automated vehicles. You can have a
fully automated vehicle that's bogged down in the network unless the network
is properly designed.

Dennis

----- Original Message -----
From: "gary" <garyd...@gmail.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Sunday, September 07, 2008 10:49 PM
Subject: [t-i] Re: PRT obsoleted by robocars

>

[Brad Templeton]

On Sep 7, 10:49 pm, gary <garydst...@gmail.com> wrote:
> This has been a fascinating dialog!  I thank Brad for kicking it off.
> In looking for info on the DARPA technologies, I found a VERY
> interesting video.  It's a full 40 minutes and starts out slow, but I
> highly recommend it to all of you...
>
> http://blip.tv/file/920488
>

I also have pointers to various other videos on the robocars.net
site. I recommend the NOVA episode "The great robot race" which is
about the 2nd Darpa Grand Challenge of 2005. It also covers the
first, which is worthwhile because you see how it went from barely
working at all, to a 150 mile desert trek with windy narrow roads and
passing other cars in just 2 years, and then if you look at other
videos, to urban streets with HDVs in another 2 years.

[Dennis Manning]

Brad:

As an innovator and pioneer in the computer/internet world and having been
an active particpant in the dazzling twists and turns of that world. I'm
surprised at how you are so certain about the path of advanced
transportation. I think it will be every bit as dazziling. Moving
information is vital. Moving people and goods is also very fundamental. You
can have all the information in the world, but if doesn't improve our
transpotration system in the movement of people and goods what's the use?

Dennis

----- Original Message -----
From: "Brad Templeton" <bra...@gmail.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Sunday, September 07, 2008 11:42 PM
Subject: [t-i] Re: PRT obsoleted by robocars

>
>
>
>

[Bruce Attah]

Brad Templeton wrote:

> People really care "Can this vehicle take me where I am going?"

They do indeed, and a guideway can take people where they want to go.
There's nowhere a conventional asphalt road can go that a guideway
can't go. It depends on the design of the guideway, of course, but the
guideway could work out cheaper than a conventional road.

> > 0.1% of trips. No consequence.
>
> Alas, not by today's buying patterns. Today people buy a car to meet
> all their needs (that they can afford.) They buy an SUV if they go to
> the hills twice a year. They buy a pickup if they haul cargo a few
> times a year. They won't buy a car with 100 miles range if they go
> 120 miles once a month.

If you're buying a car, then it makes sense to buy a car that can
serve for any trip you're likely to make by car. If the choice is not
what car to buy, but whether to buy a car at all, then the picture
changes. If you don't have to buy a car because it's cheaper and more
convenient to hire a car on a per-trip basis, you're much less likely
to buy a car for those occasional trips that you might make once or
twice a year.

> I have predicted a drop in car ownership as the need for it is
> relaxed, but I think it's much too far to predict the complete end of
> private car ownership.

I predict the nearly complete end of private car ownership if robocars
happen. It is much too conservative to predict that most people will
continue buying cars, when it makes no economic sense to do so. How
many people still buy gramophones?

> If cars are affordable and are not a burden to
> store, there will be car ownership for some portion of society that
> wishes it, and for another portion they will always hire cars. I am
> confident of this prediction for quite some time to come.

People buy cars mainly because they need them to get around. In cities
like London, Tokyo and New York, the need is less, and car ownership
is low. Nearly 80% of Manhattanites do not own a car. In New York City
as a whole, about half do without a car. If something like robocars
came along, which would presumably be quicker and much more convenient
than the New York Subway, car ownership would be lower still, and in
every city, not just the few cities that are currently well served by
public transport.

> > > First, I predict robocars will become cheaper than today's cars.
> >
> > Why?
>
> First of all, 1 and 2 person vehicles become much more marketable, and
> they actually are what you need for most trips.

One and two person vehicles are what most people need for most trips
now, but they still buy 4, 5, 6, and 7 seat vehicles. If they were
buying robocars, there's no reason to suppose the pattern would be any
different.

> Secondly, since you
> don't need long range in most vehicles (as most trips are short) the
> battery needs are far more modest. Without an expensive battery,
> electric cars are simpler and cheaper than gasoline cars, and will get
> even cheaper with mass production.

Short-range, low cost electric cars are available now. Not many people
buy them. If robocars became available, and people still wanted to own
their cars privately, they'd probably still buy cars with speed and
range comparable to the cars they today.

> Finally, once car crashes are a
> thing of the past (which is further in the future) you can build
> vehicles without all the anti-crash systems (crumple zones, roll bars,
> air bags etc.) making them much cheaper.

This is a possibility, though I suspect that the relaxing of standards
for passive stafety would lag many years behind the drop in accident
frequency.

> The computer
> part follows Moore's law and is eventually just a few hundred dollars,
> though it costs much more to start.

The computer part is never likely to be significantly cheaper than
manual control, so it's not really germane to your argument.

> > prospect that if I leave my car to pick up fares while I'm in the
> > office, I will earn more than chump change from doing so? Not great.
>
> That's why the market will balance ownership vs. hiring. You were
> saying most people will want to hire -- that makes it good prospects
> to own.

During peak hours, perhaps 50% of the population will be on the move
at one time, at most. Outside the peak hours, perhaps 5% will be on
the move at once, and the total number of off-peak trips will be less
than the total number of peak hour trips. Let's say all the peak-hour
travellers (commuters) own robocars. Let's say they all send them out
to look for fares during the day. There'll be about ten robocars
searching for each fare. In that scenario, a robocar will be lucky if
it picks up more than one fare in a day. It's hardly worthwhile,
especially if the risk of damage to the car, and the cost of dead-
heading are taken into account.

I certainly don't think people will hire out their car so they can
afford a better car. If people care that much about how nice a car
they have, and buy the nicest one they can afford, they're not likely
to be eager to hire it out. If they're entrepreneurially minded,
they'll buy the car that's best for hiring out, not the nicest car
they can afford.

> Your car will largely have a per-mile operating/depreciation cost if
> well used. If your commute is 20 miles/day, and you hire it out for
> an additional 20 miles per day, but at double your per-mile cost, it
> would in fact make a profit for you.

Don't forget the dead-heading. At the price you're suggesting, hiring
out your car would be a loss-making proposition.

> If it is possible to have a profitable robotaxi business with full
> time robotaxis, it is obviously possible to have those vehicles be
> owned by private owners who allocate them first to their own use and
> hire them out when not used.

You're forgetting about economies of scale. What's worthwhile for the
owner of a large fleet is not necessarily worthwhile for the owner of
one car.

> > > Hell, overhead power, cable and
> > > telephone lines have a bad visual impact, no matter how neat you make
> > > them, and they are just wires, they don't block the sky.
> >
> > They're ugly, nothing to do with blocking the sky.
>
> They are there in my view, and that's something people don't like.
> They also will prefer getting into and out of vehicles at ground level
> over having to ascend or descend, though of course in multi-storing
> buildings with integrated PRT station this is not a problem.

The difference between overhead cables and overhead guideways is that
the latter can be made very beautiful by the application of
architecture (viaducts and bridges are beloved of landscape
photographers), while the former are irredeemably ugly, and the only
thing to be done about them is to take them down and hide them.

[Jerry Roane]

Brad

Short response--- have to go to work
Batteries are part of the network to sip energy along the path.  This keeps the batteries very small and light weight.  If you have exposed wires someone (kid, dog or maintenance guy) will get killed.  If you have a scheme to protect the wires from the public then only the maintenance guys are exposed but they are still exposed.  By using batteries not only do we avoid these problems but we can store the solar power from the day and use it that night.  Each time you take a turn in the city you get a different fully charged battery pack that it good for the next leg of the journey.  This is all automated so the customer has no knowledge of the power delivery network.  They just drive where they want to go and there is no fuel gage. 

Imagine how much graphite dust your proposal would produce when it is not necessary.

You might check out EEstore.  They claim some amazing stuff that may change your "battery" view.

Our system is about 4.5 cents per mile but we go fast.  The cost of travel is the time wasted if you are anyone but a slacker.  You have to account for the cost of time and the 4.5 cents per mile is essentially zero in that calculation. At 4.5 cents we are well under the energy budget for solar.

German electric power "costs" the same as any other country's electric power.  They choose to jack up the rate to force solar because they are better chess players than we are and they know that draining off the nation's wealth for the next 25 years is stupid is as stupid does to steal a quote from Forrest.

Jerry Roane

[gary]

Bruce,

> The difference between overhead cables and overhead guideways is that
> the latter can be made very beautiful by the application of
> architecture (viaducts and bridges are beloved of landscape
> photographers), while the former are irredeemably ugly, and the only
> thing to be done about them is to take them down and hide them.

This issue absolutely cannot be resolved without polling people with
realistic renderings. Until then none of us will change our minds.
Beauty is indeed in the eye of the beholder.

gary

[Marsden Burger]

Bruce,

 

The Cabintaxi team backed away from bidding Detroit before the final bid.  Team management did not want to be building two "first time" city projects at one time, and the Hamburg project, which was expected to be well underway at the time of the bidding for Detroit was slipping its schedule, and creating the chance that they would have to be undertaken together.  As the schedules continued to coalesce, the decision was made to drop out of the Detroit project and build at home first.  

 

"I think it might be fair to say that a PRT guideway that
has vehicles attached both above and below is a bit more complicated
than one that supports vehicles only from above,"

 

It might be.   But the real extra cost significance is in the cost for development.  Urban integration is dramatically more effective with the over and under from both a service, operating and cost (capital and operating) point of view.  Again for those that have not seen this, here is a chart showing the systems structure cost relationships:

 

 

These cost relationships are not estimates on a future "to be developed" system .  These are real cost relationship for a finished system ready for installation.  These are true costs for real small vehicle hardware that were accepted by the entire Cabintaxi development effort.  The Cabintaxi technology was a "transparent" development by the German government with the over site and involvement of five public transit and rail organizations which literally play a significant role in setting the world standards for fixed rail safety - the German Federal Railway (DB), the Technical Examination Authority (TUV), the Association of Public Transportation Operators (VOV), the Industrial Plant and Operation Advisory (IABG), the Hamburg Elevated Rail system (Hamburg Hochbahn) - not to mention the German Ministry for Research and Technology.  Nothing in transit development history has come close to the program developed in Germany in the 70's, and ongoing with Transrapid to this day.  In my opinion, the closest to this amazing effort, was the equally amazing individual effort of Kirk Foley and his team in the creation of the Ontario Transportation Development Corporation in Canada, which also had the wisdom (and the funding) to involve the true transportation world in the development effort.

Notice above that the three passenger over running system is effectively the same as any over running three passenger vehicle system concept in the world, again for an "urban" PRT system - not dual mode, not intercity, not etc..... Can a system be developed that will have a less expensive guideway - probably, but not much less.  However, the issue is between conventional rail and small vehicle systems; the difference between a Cabintaxi system costs and other yet undeveloped three passenger systems is meaningless in the comparison to conventional rail to small vehicles.

I think it is also important to note that in all of the simulation work done for application of small vehicle systems in Germany, undoubtedly more real world detailed simulation work done than anywhere else in the world on this subject, (examples shown below) I am unaware of a study that showed a single level PRT system found to be more effective than the over and under.  Had an application been better served by the single level approach, the Cabintaxi effort would have supplied it - we had two single level PRT systems. 

Also note above the cost relationship differences between an under running single level system and an over running single level system.  Is there a difference - sure.  Is it worth arguing about?

The important issues for our advanced technology efforts are how do we get major transportation improvements that can change the quality of life for mankind into installation.   

 

 

 

 

 

 

 

 "an APM system that also offers PRT is more complicated than a pure
APM."

 

 However, a PRT system that is already cheaper than an APM (in Detroit, our APM competitors were running standing passenger vehicles, while our bid approach, approved by the customer, was to run 12 passenger seated passenger vehicles) becomes cheaper yet when switching to to APM (but still seated) style of operation.  And, if desired, that change can be a simple as pushing a button.  The problem with other PRT system having this capability, is the cost of development, where they are having trouble finding the development funds to do the efforts already on their plates.  It is the cost and testing of full scale vehicles on full scale guideway, and then refining the results, that is most expensive.  This is the power of the $270 million (today's dollars) of the Cabintaxi development.

 

"In another application, where
a simple one-way guideway provided adequate capacity, I imagine that
the Cabintaxi system would tend to be somewhat more expensive (perhaps
only slightly). Is this fair, or am I wildly off base?"

Here again, if there was such a scenario, the Cabintaxi technology could simply supply a single level PRT system.  However, I must let my interest in providing effective transportation service push the aspects of PRT technology out of the way.  If a one directional PRT guideway provided adequate capacity assuming that a larger network where higher capacity and greater service is not in the offing, then the probability is high that a much less complicated group rapid transit system, still utilizing small seated passenger vehicles of 12 passenger or 18 passenger size, in a radically smaller fleet, less complex and costly stations - no off lines, and a radically cheaper maintenance facility, would undoubtedly provide the service required.  A classic example of this is that Morgantown, at the push of a button, can run as a PRT operation, but chooses to more effective, over its guideway and in its geography, run as GRT. 

 

Look again at the guideway cost relationship in the above chart.  Two single lanes of one level guideway running in a dual track configuration are cheaper to connect a small application linearly than turning that linear connected dual lane system into a one way loop, in most applications in the real world the loop will require more guideway feet.  Again, add to this the extra complication of switches, more complex controls, the extra length of guideway needed for off line stations, the low lane capacity challenged by a rush, the high cost of the maintenance facility to handle a larger number of vehicles, and you start to get the understanding of why starting a PRT system from scratch is a hard thing to accomplish.

 

The recognition of the problem of initiating a full PRT system, and the significant initial service potential possible by applying higher capacity small vehicles over light guideways, was what prompted the Cabintaxi development effort to expand the vehicle types to the 6, 12, and 18 passenger vehicle concepts.  Cabintaxi can be effective as a simple shuttle to start, and expand to become a full city wide PRT system if required.  Again, the critical costs to make this available are development costs.  They are successfully done and paid for with the Cabintaxi technology.

 

 

 

 

    "PS, by "perfectly", I of course did not mean "unimprovably" -- merely
that it did what it said on the tin."

 

I know.  My "Smiley Face" at the end of my sentence for some reason did not show up."

 

Best wishes,

 

Marsden

 

----- Original Message -----

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "transport-innovators" group.
To post to this group, send email to transport-...@googlegroups.com
To unsubscribe from this group, send email to transport-innova...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en
-~----------~----~----~----~------~----~------~--~---

[Dennis Manning]

Gary:

Thanks for posting. Yes, it was interesting. The sensor technology has
certainly taken a leap. I noticed the prediction as to when drivers would
finally let go of wheel was about 20 years out, but a much shorter time for
more driver aides to be used.

When I rode in the buicks in the 1996 demo of platooning cars in San Diego
the equipment took up most of the trunk.

Dennis

----- Original Message -----
From: "gary" <garyd...@gmail.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Sunday, September 07, 2008 10:49 PM
Subject: [t-i] Re: PRT obsoleted by robocars

>

[gary]

That presentation also did a good job IMO of addressing the liability
issue. For the near term future this technology will be effectively
an advanced cruise control, so the driver will STILL be liable (and
behind the wheel). But at some point the automation will be safer and
it will be the INSURANCE companies that insist on their usage over
manual control, thus flipping the liability issue.

I firmly believe that robocars will be technically possible in the
future, but still believe rail is a better approach (safer, always
connected to grid, more energy efficient, rails less expensive to
maintain than asphalt, and so on). Robocars might be a good choice
for those adventerous folks who want to head camping, but for the
other 99.999 percentage of miles traveled, the path is well worn. And
thanks to DARPA's open contract approach to the challenges, the
necessary collision avoidance technology is even further along than I
realized. I got a lot of flack early on for suggesting a ground based
system could operate safely. I wonder if anyone who's watched the
above mentioned video still maintains that position.

gary



On Sep 8, 12:03 pm, "Dennis Manning" <john.manni...@comcast.net>
wrote:

> > gary- Hide quoted text -
>
> - Show quoted text -

[Brad Templeton]

On Sep 8, 12:08 am, "Dennis Manning" <john.manni...@comcast.net>
wrote:

> Brad:
>
> As an innovator and pioneer in the computer/internet world and having been
> an active particpant in the dazzling twists and turns of that world. I'm
> surprised at how you are so certain about the path of advanced
> transportation. I think it will be every bit as dazziling. Moving
> information is vital. Moving people and goods is also very fundamental. You
> can have all the information in the world, but if doesn't improve our
> transpotration system in the movement of people and goods what's the use?
>

Certain? Far from it. I outline on my web site a score of things
that can and will go wrong, as well as various downsides.

What I am fairly sure of is that it's *possible* to do it, if we have
the will. And further, it's fairly clear to me that it's well worth
doing, because of the lives it will save, the time and money it will
save, and the energy it can save.

So my focus is twofold. What's the path that might make this
happen? And when it happens, what does it mean for transportation,
society, cities, manufacturing, energy and many other factors.

It's truly world-changing, in a way no other transportation innovation
I have seen is.

I've become more confident it can happen soon -- if we, or some other
country, have the will -- by the highly positive results of
prototypes. It is easy to be fooled by prototypes, that is to be
sure (PRT people should know that better than anybody) but they are
far enough along it's worth considering the factors. For those who
say it can't be done ever, I have little time. For those who say
"It's going to take a very long time" there is worthy debate.

But now let me rephrase a question you asked

"If you have a vehicle that is computer controlled then what is the

best path/guideway?" You said you didn't think it was roads.

What if I ask it as "If you have a vehicle that can operate safely on
a flat surface with no need of a guideway, what is the best path/
guideway?"

I find it quite surprising to see people answer that question with
other than roads, from a pragmatic sense. Yes, I can see somebody
advocating rails or other guideways if we were building a city from
scratch. They might be cheaper to build or operate. But when we
already have a giant network of flat surfaces that is built and goes
everywhere (including dirt tracks up the mountains,) and any new rails
or guideways would have to be built from scratch and would not go
everywhere even after many decades, what other answer can compete?

Especially when you consider the other advantages of flat surfaces

a) They can be arbitrary widths and automatically handle multiple
vehicles in parallel
b) They can handle any size of vehicle they are wide enough for, from
a two-wheeled bicycle to a 32 foot wide Tar Sands mining truck, to new
vehicles not yet dreamed of
c) People can walk over them easily, and as noted, bike on them
d) If there is an obstacle or breakdown, vehicles can just move around
it.
e) They require almost no standardization (which limits innovation) of
the vehicles which travel on them
f) When you are ready to ban human drivers from portions of them (as
you inherently do on guideways) they can act as guideways as far as
traffic patterns are concerned, the walls are just virtual instead of
physical.
g) If you want to build new ones that are elevated and only take
narrow-width robocars, and were even electrified, you certainly
could. But I doubt you would bother.


But most of all, they already exist, and will exist and be maintained
for many decades to come. Rails and guideways can offer us only
the slightly better rolling resistance of steel/steel, and slightly
easier electrification. But can these advantages outweigh all that
is above?

[Brad Templeton]

> They do indeed, and a guideway can take people where they want to go.
> There's nowhere a conventional asphalt road can go that a guideway
> can't go. It depends on the design of the guideway, of course, but the
> guideway could work out cheaper than a conventional road.

Really? Out on the farm? On the playa at burning man? Cheaper than
a gravel driveway as many rural people have? Would fire trucks ride
the guideways? If we don't have guideways everywhere -- and even in
the optimistic world this would not happen for a very long time --
what sort of fire trucks and ambulances will we have? And what will
they ride on? We would need both kinds, of course, meaning both
networks must be fully connected -- ie. you can' t have an island of
roads that can't reach other roads unless you give it a full set of
emergency vehicles.

> People buy cars mainly because they need them to get around. In cities
> like London, Tokyo and New York, the need is less, and car ownership
> is low. Nearly 80% of Manhattanites do not own a car. In New York City
> as a whole, about half do without a car. If something like robocars
> came along, which would presumably be quicker and much more convenient
> than the New York Subway, car ownership would be lower still, and in
> every city, not just the few cities that are currently well served by
> public transport.

Even in Hong Kong, where car ownership is crazy, and car trips are
slower than transit in the core, people still own and use cars.

Anyway, the market will figure out if people want to own cars or not.
It would be incorrect to design based on a prediction that people will
stop wanting to own cars, and incorrect to design based on a
prediction that they will. Cars will be for sale, and while you might
predict only taxi companies would bother to buy them to hire out,
that's really up to the people. They will buy them if they wish.
Since a large fraction of today's homes have garages, and I predict
robocars will be cheap to buy and cheap to store, I am confident that
some people will buy them.

This will also be affected by how large the cloud of available
vehicles is. If you live somewhere more rural, where the wait time
for a car is longer, you may wish to own for that reason.

what you need for most trips.
>
> One and two person vehicles are what most people need for most trips
> now, but they still buy 4, 5, 6, and 7 seat vehicles. If they were
> buying robocars, there's no reason to suppose the pattern would be any
> different.
>

Have you read my essays? I go into this in detail. The key factor
is what I thought you believe in. If you can hire a robocar for your
rare needs (such as taking 5 people) you may still want to own a car,
but there is less pressure to get a car that handles all your needs,
rather than your main needs. So it makes perfect sense to buy a 1
person car for your commute, and to summon a hired minivan when taking
the whole family out. Today nobody buys 1 person cars because they
don't want a car that only does one thing.

>
> Short-range, low cost electric cars are available now. Not many people
> buy them. If robocars became available, and people still wanted to own
> their cars privately, they'd probably still buy cars with speed and

> range comparable to the cars they today.'

I am getting the sense you are debating me without having read my
essays. robocars.net points to them.

> During peak hours, perhaps 50% of the population will be on the move
> at one time, at most. Outside the peak hours, perhaps 5% will be on
> the move at once, and the total number of off-peak trips will be less
> than the total number of peak hour trips. Let's say all the peak-hour
> travellers (commuters) own robocars. Let's say they all send them out
> to look for fares during the day. There'll be about ten robocars
> searching for each fare. In that scenario, a robocar will be lucky if
> it picks up more than one fare in a day. It's hardly worthwhile,
> especially if the risk of damage to the car, and the cost of dead-
> heading are taken into account.

First of all, not all peak commuters will own, because there will be a
hirable fleet out there, consisting of full-time taxis and the
vehicles of people who don't commute at the peak time. Which is
quite a few people, actually, we've gotten much more staggered in
response to congestion and other factors.

For the people doing short, urban commutes of a few miles (quite a
lot) it becomes practical, if not very energy efficient, to have cars
return empty to bring in
another commuter (which trains and buses often have to do.) This is
less practical for suburban commutes, which will push more ownership
in that area.

(In the long term, I expect the "erasure of distance" that robocars
bring to slowly redesign downtowns to be less concentrated, allowing
more efficient
two-way commutes, but this is a much less certain prediction.
Because a robocar is also a working/leisure environment -- with
internet, screen, keyboard, phone etc. --- people will view some trips
as not wasting any time at all, unlike today's trips.)

>
> I certainly don't think people will hire out their car so they can
> afford a better car. If people care that much about how nice a car
> they have, and buy the nicest one they can afford, they're not likely
> to be eager to hire it out. If they're entrepreneurially minded,
> they'll buy the car that's best for hiring out, not the nicest car
> they can afford.

It depends on their income. A very simple "hire out" would just be an
agreement to share a pool of cars among a group of associates. Where
today 10 families might have 20 cars, the 10 families could probably
do fine on far fewer, especially if their commutes are complementary.
And if not, they can always temporarily hire another car for spot
needs.

>

>
> Don't forget the dead-heading. At the price you're suggesting, hiring
> out your car would be a loss-making proposition.

The market will figure this out. There will be demand for hired
cars, and there will be a market price. The price will have to be
high enough that owners, be they private or taxi companies, find it
profitable. What price that is, I don't know, but I know it
exists. Including vacant trip costs, of course. I expect
sophisticated software will be developed to optimize fleets and reduce
such trips.

>
> You're forgetting about economies of scale. What's worthwhile for the
> owner of a large fleet is not necessarily worthwhile for the owner of
> one car.

Actually, quite the reverse. Computers are good at aggregating things
and making it scale.
I anticipate an owner would actually contract with a taxi company to
work as intermediary. The taxi company would look at customer
requests, and if a privately owned car under contract has registered
itself as available and nearby, it will use it at a negotiated price
-- possibly a price negotiated that second by the computers.

That's what computer nets are good for. The mods to make your car
workable as a robotaxi are not too great -- mainly a before/after
camera and an external display if you don't already have one. And you
must keep your personal stuff in a locker rather than scattered about
your car. Some people will not want to do that and they won't hire
out. Others will. Today most cars are sitting doing nothing most
of the time. We can make this more efficient.

[gary]

Brad,

I think I might be up your way next week...can I rent your car? I
hope you don't mind if I smoke. We're bringing the great danes to the
countryside to go for a swim. And of course we're bringing little
Sally as well. We've found we can pacify her on those windy roads
with a bag of cookies, although she sometimes has a hard time keeping
them down. Anyway, let me know when I can come pickup the keys...

humorously yours,
gary

[Dennis Manning]

Brad wrote (portion)

<But now let me rephrase a question you asked

"If you have a vehicle that is computer controlled then what is the
best path/guideway?" You said you didn't think it was roads.

What if I ask it as "If you have a vehicle that can operate safely on
a flat surface with no need of a guideway, what is the best path/
guideway?">

That's not a question since you state "no need of a guideway". So I have
your answer.

Dennis

[Brad Templeton]

On Sep 8, 2:06 pm, gary <garydst...@gmail.com> wrote:
> Brad,
>
> I think I might be up your way next week...can I rent your car?  I
> hope you don't mind if I smoke.  We're bringing the great danes to the
> countryside to go for a swim.  And of course we're bringing little
> Sally as well.  We've found we can pacify her on those windy roads
> with a bag of cookies, although she sometimes has a hard time keeping
> them down.  Anyway, let me know when I can come pickup the keys...
>
> humorously yours,
> gary

That's why you want intermediary companies handling it. You act as
though nobody would ever rent a car. Car rental companies do, and the
cars are always clean and ready when I rent them, no matter who was in
them before with little kids etc. (Some car rental companies do ban
smoking, as I would.)

I anticipate any rental robcar will have a camera/sensor which scans
the car before any new renter enters it, and after they leave. (It
would be physically shuttered while in use for privacy reasons.) If
there's a difference between the two scans, a human would look at them
to see what the problem is. Most of the time, it would be something
left in the vehicle, and the renter would be alerted to this before
they got 20 feet from the car on their cell phone.

If the renter left a mess, the car would take itself to a cleaning
station -- that's the great thing about robocars, they move themselves
-- and the renter would be billed for it. If the renter did damage
that could not be fixed their insurance would pay for the major
repair.

Now with robocars, your car could be rented with you literally doing
nothing but agreeing to a contract, and making it available when you
aren't using it. That's what's new -- no logistics for you to worry
about, few for the taxi company.

So sure, under those terms, I would be happy to have my car make money
for me while I am not using it, if I'm the sort of person who wants
the income (as many do) and who keeps their stuff in a box rather than
scattered over the trunk. Remember that people here are declaring
nobody will want to own a car at all. If you don't own any car you
can't keep your stuff in one, so this is not much of a leap.

But again, this is a matter of individual taste. Some people won't
own cars at all. Some will own them and never let anybody touch
them. Some will share cars with a small group of trusted friends and
neighbours. Some will be happy to hire them out for income.

I think it might be enough in fact that taxi companies own few cars.
Instead they would broker, renting out spare time on cars owned by
private individuals. This is how an RV rental company near me that I
have used works. They own no RVs. The private owners store their RV
at the rental place, and the burden is they have to reserve their own
RV in advance, they can't just go grab it when they want as it might
be out on rental. And they can't keep much stuff in it. The rental
house handles all logistics, assures the RV is cleaned after each
rental. The owners actually make money, especially if they mostly
use their RV in the off season and let renters have it in the high
season. They get a free RV, effectively. It works, and doesn't even
have robot driving. I'm surprised you folks are so skeptical of it.

There is so much spare car capacity in the world. I have not seen
figures, but I bet that fewer than 1/4 of all cars are on the road
even at the peak of rush hour, and outside of rush hour, I would bet
less than 5% of cars are on the road. Considering that the energy
required to make a car is 25% of lifetime energy for a Prius, more
like 10% for more ordinary cars, this is huge.

[Brad Templeton]

Well, the strange thing was, I seemed to be getting people here saying
that even with robocars (which, as I define them, have no need for a
guideway) it would be better to build guideways than run the robocars
on existing streets. That's the thing I can't figure.

[Jerry Schneider]

At 04:47 PM 9/8/2008, Brad wrote:
>snip -------------------------------

>There is so much spare car capacity in the world. I have not seen
>figures, but I bet that fewer than 1/4 of all cars are on the road
>even at the peak of rush hour, and outside of rush hour, I would bet
>less than 5% of cars are on the road. Considering that the energy
>required to make a car is 25% of lifetime energy for a Prius, more
>like 10% for more ordinary cars, this is huge.

If the annual average mileage per auto is about 15,000 miles/year,
dividing by 365 days gives
an average of 41 miles per day - or about 1.36 hours per day per
vehicle, at 30 mph, which is
5.6 % of 24 hours - with lots of variation around the mean. Of
course, time of day, day of week,
season, etc. does affect what's "on the road" at any particular time
or time period.

I did a research project once that attempted to calculate where the
population of a city was located, stationary or
in motion, during the day. It was funded with earthquake research
money, These people make the assumption
that everyone is home all the time (i.e. the nighttime population
distribution) when making their damage forecasts
for an earthquake of a particular magnitude and epicenter location.
My Chinese student developed a computer program that did reams of
calculations (using some heroic assumptions) that showed how far off
this assumption is likely to be during different times of the day.

- Jerry Schneider -
Innovative Transportation Technologies
http://faculty.washington.edu/jbs/itrans

[Bruce Attah]

Brad Templeton wrote:
> > It depends on the design of the guideway, of course, but the
> > guideway could work out cheaper than a conventional road.
>
> Really? Out on the farm?

Guideways could be cheaper than asphalt roads, regardless of where you
put them. It depends on the specific design of the guideway. The ULTra
guideway, for instance, is just two concrete planks plus kerbs, laid
on a flat surface. It is in fact cheaper than an asphalt road. The
majority of people in developed countries rarely or never take their
cars off-road -- even people who buy off-roaders. If the robocar is
catering to the tiny few who do, it is catering to a niche market.

> Even in Hong Kong, where car ownership is crazy, and car trips are
> slower than transit in the core, people still own and use cars.

In Hong Kong, people still own and use cars, but how many? It turns
out that the number is 57 people out of 1,000. Fewer than 6%.
Singapore is another relevant example, come to think of it. There, 103
out of a thousand own cars.

People in general are more rational than you seem to think. They will
not automatically buy cars if there is no practical advantage in doing
so.

The arrival of robotaxis will mean there is no practical advantage in
car ownership for the average person.

> Anyway, the market will figure out if people want to own cars or not.

Indeed it will.

> It would be incorrect to design based on a prediction that people will
> stop wanting to own cars, and incorrect to design based on a
> prediction that they will.

I fail to see how the long-term prediction makes any difference to the
short term design of robocars (or robotaxis, as they will inevitably
turn out to be, regardless of original intention).

> Since a large fraction of today's homes have garages

A garage is just a gym or hobby room that is waiting to find its true
purpose.

> I am confident that some people will buy them.

Key word: some.

> This will also be affected by how large the cloud of available
> vehicles is. If you live somewhere more rural, where the wait time
> for a car is longer, you may wish to own for that reason.

Naturellement. But that's a small minority of people, and getting
smaller every year.

> So it makes perfect sense to buy a 1
> person car for your commute, and to summon a hired minivan when taking
> the whole family out.

It makes no sense to buy a 1 person car for your commute if robotaxis
are in abundant supply. The commute will be cheaper by taxi, so why
bother owning a car?

> I am getting the sense you are debating me without having read my
> essays.

I have read at least some of your essays, you will be gratified to
know. If I have missed any specific thing, I would prefer a link to
the specific thing I am supposed to have missed, rather than a vague
reference to your essays as a whole.

> First of all, not all peak commuters will own,

Indeed, the fewer people own robocars, the more potential profit there
is in hiring them out. If nearly everyone owns one, there's no point.

> A very simple "hire out" would just be an
> agreement to share a pool of cars among a group of associates. Where
> today 10 families might have 20 cars, the 10 families could probably
> do fine on far fewer, especially if their commutes are complementary.

That sounds to me like a recipe for quarrels.

> > You're forgetting about economies of scale. What's worthwhile for the
> > owner of a large fleet is not necessarily worthwhile for the owner of
> > one car.
>
> Actually, quite the reverse. Computers are good at aggregating things
> and making it scale.

How are computers supposed to aggregate the cleaning of the car
interior after a mucky fare? Fleets can devise economical ways to do
this sort of thing, but individuals would probably find it harder.

> I anticipate an owner would actually contract with a taxi company to
> work as intermediary. The taxi company would look at customer
> requests, and if a privately owned car under contract has registered
> itself as available and nearby, it will use it at a negotiated price
> -- possibly a price negotiated that second by the computers.

That seems to me to be pretty pointless. The taxi company is unlikely
to need to hire vehicles from individuals (bearing in mind that the
availability of those privately owned cars is subject to the whim of
the owner), and even if this arrangement were to come about, the
individual would probably end up with negligible earnings after the
taxi company had taken its cut.

These models you suggest do not strike me as at all credible.

> Today most cars are sitting doing nothing most
> of the time. We can make this more efficient.

And the most straightforward way to do this is to forget about private
cars, and leave the hassles of vehicle ownership and maintenance in
the hands of fleet operators.

[Bruce Attah]

Brad Templeton wrote:

> Well, the strange thing was, I seemed to be getting people here saying
> that even with robocars (which, as I define them, have no need for a
> guideway) it would be better to build guideways than run the robocars
> on existing streets. That's the thing I can't figure.

Some situations where a guideway might make more sense than a
conventional road:

1) you want very long range without fuel stops, so you provide a
guideway with built-in power supply.

2) you want to go very fast, and you want to save vehicle weight by
not having the motors in the vehicle, so you provide a guideway with
embedded linear motors.

3) you want to save money on road building and maintenance, so instead
of a full-width road, you supply a guideway that consists merely of
two narrow tracks for wheels.

[Brad Templeton]

>
> Guideways could be cheaper than asphalt roads, regardless of where you
> put them. It depends on the specific design of the guideway.

Of course. The problem is it's so specific. Once you decide on a
guideway tech, you are going to be stuck with it. If you decide you
really wanted a different kind of guideway (or a road) there's a huge
cost of changing. Every PRT system I see proposed has a different
sort of guideway, and arguments about why it's best/cheapest/nicest
looking/easiest to maintain etc.

A lot of history suggests that locking in to one system is a
mistake. (Though the stories I have heard about why BART decided to
go wider gauge on their rail are amusing if true.)

As people want to keep reminding me (and I know very well) we don't
actually know what the future will bring us. We can only speculate
and extrapolate. If you want the whole world to go on guideways, you
had better know that the choice was right.

It's one of the reasons that roads won over railroads, even though
railroads are much more energy efficient. Anything can go on them.
Cars, bicycles, motorcycles, trucks, road trains, experimental solar
vehicles, you name it.

And guideways have another big issue if something gets stuck on one.

The ULTra
> guideway, for instance, is just two concrete planks plus kerbs, laid
> on a flat surface.

Right. ULTra is really a not-quite-robocar. It runs on a road, on
tires, and today needs those guides at the side to keep it in line,
but something like it won't need those soon. Though for backup I
would want them on an elevated track where slight failure is death
rather than fender-bender as it is on the ground.

I'm surprised you bring it up, I think it demonstrates the versatility
of roads and tires, not of guideways and rails.

> In Hong Kong, people still own and use cars, but how many? It turns
> out that the number is 57 people out of 1,000. Fewer than 6%.
> Singapore is another relevant example, come to think of it. There, 103
> out of a thousand own cars.

Correct. That's my point, even where it's silly to have cars, people
still want them, even in a totally non-car culture. The USA is 50
years from that much of a cultural rewrite, if ever.

> Naturellement. But that's a small minority of people, and getting
> smaller every year.

This is something I wonder about. Will robocars encourage people to
live urban, where the whole city is at your disposal with a short
robocar ride, or will they encourage people to live suburban, where
everything you buy can be delivered by robot, and your trips, while
longer, are work/watch-TV/read time rather than wasted time.

It will be a mix, but we don't know enough yet to figure the answer.

> are in abundant supply. The commute will be cheaper by taxi, so why
> bother owning a car?

I realize you think this. Just accept it's your opinion of tastes.
Other people have given me very different opinions of tastes. I
predict there will be a plethora of tastes, and we don't know enough
yet to figure out the balance. So I'm not really interested in
further debate. Especially since I usually spend my time arguing with
people saying the reverse of you, that they can't imagine people not
owning cars, and I have a number of proposals to help them imagine it
(like stuff lockers and car clubs) but the truth is there are strong
opinions in both directions. Opinions. The reality we will have to
see when it comes.

>
> How are computers supposed to aggregate the cleaning of the car
> interior after a mucky fare? Fleets can devise economical ways to do
> this sort of thing, but individuals would probably find it harder.

It's outlined in my "week of robocars" among other places. When a
car is detected to be dirty (probably by the computer) it takes itself
to a contracted cleaning station. The party who dirtied it pays.
The owner may not even be aware this took place.

> That seems to me to be pretty pointless. The taxi company is unlikely
> to need to hire vehicles from individuals

Quite the reverse. Why should the taxi company wish to buy vehicles
if car owners are willing to supply it with vehicles that, since they
barely use them, they will hire out for less? Again, the market
will figure out these things. The cost of owning a car by a taxi
company. The cost of owning a car for an individual. The minimum
price various classes of individual owners will hire out for. etc.

If I were a taxi company, I would examine trends in availability of
contracted privately owned vehicles, and only buy enough cars to meet
customer demand that can't be met by that supply.

I also predict a spot market. For example, some owners will say,
"While the ordinary rate is 10 cents/mile, I will only rent out for 40
cents/mile because I am richer or don't like renting so much." The
taxi companies, at peak demand times, would pay this 40 cents (losing
money on the fare, probably, but keeping their promise to clients of
always having a vehicle for them.) This is a common strategy for
peak demand management, pay extra only at the peak.

(bearing in mind that the
> availability of those privately owned cars is subject to the whim of
> the owner), and even if this arrangement were to come about, the
> individual would probably end up with negligible earnings after the
> taxi company had taken its cut.

Again, what matters to the owner is are they paid enough. Anything
above that, less depreciation for a car they weren't going to use is
found money to the owners. Everybody has their price.

>
> These models you suggest do not strike me as at all credible.

We don't have enough info to predict or make a statement like that.

[Walter Brewer]

I believe the auto use average is ~ 6 trips/day; average trip length ~6
miles. That checks your 41 mile number.

Walt Brewer

----- Original Message -----
From: "Jerry Schneider" <j...@peak.org>
To: <transport-...@googlegroups.com>
Sent: Monday, September 08, 2008 8:30 PM
Subject: [t-i] Re: PRT obsoleted by robocars

>

[gary]

Brad,

As our vehicles become more and more complex, it makes less sense for
people to own them as you run the risk of "modifications" that cause
accidents. Or vehicles that don't get serviced when they should.
Another advantage of public ownership is that the total number of
vehicles requires is much smaller (a VERY green advantage). And you
can get back your garage. And driveway. And all those wasted, ugly
parking lots at every store, school, workplace, and public building in
town. And without private ownership, you wouldn't need all the silly
gratitutous car models we currently have. You'd just have the 2-
seater, the 4-seater, etc. As for your "rent my car while I'm not
using it" idea, I can foresee all sorts of problems...it's not back
when you need it, fights over trash left in the car, etc. Who need
the headache? I just don't think anyone would go through all that
effort. But like you said, marketing forces will prevail, so no need
to worry about it. In my scenario (publicly owned vehicles) I have a
similar scheme for dealing with dirty vehicles. You just tag it for
cleaning, jump in the replacement, and off you go. That's all you
need to worry about. Behind the scenes the system will of course
detect repeat offenders, verify via video, and take appropriate
action.

So in conclusion I want the same luxury that Donald Trump has. He
simply summons a vehicle, it picks him up at the front door, he fires
a few employees over the phone to relax, and he gets out right at his
destination. No thinking about what size vehicle to buy, getting the
vehicle serviced, buying insurance, renting it out, or any of the crap
that goes with ownership. Doesn't that sound a lot easier?

gary

[Brad Templeton]

On Sep 8, 9:09 pm, gary <garydst...@gmail.com> wrote:
> Brad,
>

> As our vehicles become more and more complex, it makes less sense for
> people to own them as you run the risk of "modifications" that cause
> accidents.  Or vehicles that don't get serviced when they should.

Sure, you can come up with all sorts of advantages for public
ownership, or rather in this case, private fleet ownership. I
understand them all, but they are all dwarfed, *dwarfed* by the big
advantage of private ownership. Innovation. A bustling market of
entrepreneurs dreaming up products to sell to individuals, with vast
choice and competition, and people buying expensive new things because
they are cool -- that's where innovation comes from. It's not
impossible in other situations, but this is where it really thrives.

So you get your choice. Public ownership of old technology with all
the "advantages" of public ownership, or private ownership with all
its disadvantages, of something far better.

You're in the timesharing/mainframe world. I'm in the PC world.
You're in the Cable company set top box world, I'm in the Tivo/MythTV
world. You're in the licenced microwave data link world, I'm in the
wifi world. You're running Windows, I'm running linux. (I don't
mean you, I mean your approach, of course.) You're driving a Lada,
I'm driving a Prius.

> Another advantage of public ownership is that the total number of
> vehicles requires is much smaller (a VERY green advantage).

It is greener, but I think the market will drive things in the green
direction if the economies of the energy used are properly expressed
in the price. People don't go green because you tell them it's a
nice thing to do. Other than a small minority, they go green when
it's cheaper. So in fact, I predict that many people will elect not
to own, and many people will elect to own but hire out their cars as a
side business, and this will get many of the benefits you are looking
for. Yes, people will want their garages back. Yes, people will
want to build condos on their old parking lots. The market will push
things this way.

 And without private ownership, you wouldn't need all the silly
> gratitutous car models we currently have.

You write like that's a bad thing. :-) Seriously, consumer choice
is not a bad thing, just because you don't like where it leads.

..it's not back
> when you need it,

So rent one yourself. Or insist that the person using it longer than
you said it was available get out and switch to another.

fights over trash left in the car, etc.  Who need

> the headache?'

What fights? It will be very obvious who does this in the system I
have described. What headache? The computer sees the car looks
different inside than when the passengers got in, and it goes to a
cleaning depot. They bill the passenger, including your cost to rent
a temporary replacement.

 In my scenario (publicly owned vehicles) I have a
> similar scheme for dealing with dirty vehicles.  You just tag it for
> cleaning, jump in the replacement, and off you go.  That's all you
> need to worry about.  Behind the scenes the system will of course
> detect repeat offenders, verify via video, and take appropriate
> action.

What do you refer to as a publicly owned vehicle? You don't mean the
government I hope. I am presuming you mean a taxi company. Why is a
taxi company so special an owner, compared to a private individual who
lets a taxi company handle the logistics?

>
> So in conclusion I want the same luxury that Donald Trump has.  He
> simply summons a vehicle, it picks him up at the front door, he fires
> a few employees over the phone to relax, and he gets out right at his
> destination.  No thinking about what size vehicle to buy, getting the
> vehicle serviced, buying insurance, renting it out, or any of the crap
> that goes with ownership.  Doesn't that sound a lot easier?

Indeed, I suspect many will choose that. In fact, if you read my
essays, it's even better. The rental car comes to you digitally
configured for you -- your phone, your web configuration on the
computer, your slogan on the bumper, the seats adjusted for your body,
the level of luxury you paid for and -- if you gave enough advanced
warning -- a lockbox in the trunk with "your stuff". I have
designed proposals to make it as palatable as possible not to own.

And yet, the more people I talk to and explain this to, the more I
know some people will still want to own. Talk to more folks, describe
this vision, and you'll see. The shift towards New York thinking
(let alone Hong Kong thinking) will take a very long time, if it ever
finishes.

[Dennis Manning]

Brad:

I think I'll let you and gary duke it out on all the robocar speculations.

My thoughts run along the lines that the future (20-50 years out) will
contain manually controlled cars and other manually controlled vehicles, and
robocars, and automated fixed guideway systems (networks). The questions are
market share and timing. All of the systems we discuss are already in the
mix. Even robocars and PRT have their infant renditions. What I think is
missed in much of this discussion is each system has it's merits, it's
niche. PRT can do things that robos can't and vice versa.

Arguably the earliest automated fixed guideway system in wide use is the
elevator. Then came APMs (horizontal). Then PRT functionality is realized at
Morgantown and proved doable by Cabintaxi and Raytheon (network). The single
biggest advantage of PRT over robocar may eventually be the right of way
competition.

Robo vehicles putter around hospitals, warehouses, and large scale vehicles
at ports.

We will just see what progresses and how soon.

The ULTra and FROG systems are sort of hybrids. They run on a flat surface,
steer themselves, have obstacle sensors, etc. What the difference is between
the fork in the road leading to gary's or Brad's vision is the degree to
which the path is separated from existing traffic.

One more item. At grade systems must compete with what's already using the
space. There isn't a lot of competition for space 25 feet above grade (maybe
some buildings excepted).

It depends on where you are that determines whether at grade or elevated is
the most expensive.

Dennis

----- Original Message -----
From: "Brad Templeton" <bra...@gmail.com>

To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Monday, September 08, 2008 7:14 PM
Subject: [t-i] Re: PRT obsoleted by robocars

>
>
>
>>

[rob...@2getthere.eu]

Bruce,

Concrete is a difficult underground for a transport system. It's ride
comfort charactaristics are below par and it is hard to maintain
properly. The seems between the concrete will cause the sensation of a
train. Asphalt is really unbeatable with regard to those aspects. In
capital costs there is not much difference, if you compare apples with
apples (obivously there is if you compare concrete 'tracks' with a
complete asphalt road).

Robbert

[Mike C]

On Sep 9, 12:35 am, Brad Templeton <brad...@gmail.com> wrote:
> On Sep 8, 9:09 pm, gary <garydst...@gmail.com> wrote:
>
> > Brad,
>
> > As our vehicles become more and more complex, it makes less sense for
> > people to own them as you run the risk of "modifications" that cause
> > accidents.  Or vehicles that don't get serviced when they should.
>

> You're in the timesharing/mainframe world.  I'm in the PC world.
> You're in the Cable company set top box world, I'm in the Tivo/MythTV
> world.   You're in the licenced microwave data link world, I'm in the
> wifi world.   You're running Windows, I'm running linux.   (I don't
> mean you, I mean your approach, of course.)   You're driving a Lada,
> I'm driving a Prius.

I think some of your analogies are wrong. The PC may rule today, but
our computer activities are mainly in the shared, public Internet. You
may have a Tivo, but you now "own" less of your entertainment,
choosing to record/watch/delete using the on-demand features of Tivo.
Wifi is a *shared* system. Linux may be popular among a certain
minority, but most people prefer Macs and Windows.

Look at the entertainment world: between NetFlix, on-demand cable/
satellite, and now online video, we "rent" a majority of our
entertainment now. The things we buy are throwaway, like $1 for a a
bunch of bits at iTunes. Look at what the Internet did to AOL and
Compuserve - both "private" systems that were smothered by the public
Internet.

>
> And yet, the more people I talk to and explain this to, the more I
> know some people will still want to own. Talk to more folks, describe
> this vision, and you'll see. The shift towards New York thinking
> (let alone Hong Kong thinking) will take a very long time, if it ever
> finishes.

My feeling is that people own cars because that's what you have to do
today to have the freedom of a car. They buy the freedom, not the
hardware. If you give them freedom of movement in a non-owned system,
they will accept it just as openly as they do their privately owned
automobile (there will be a few exceptions, of course, like those car
owners in Hong Kong, but I'm talking about trends here).

I think the people who say otherwise basically don't understand what
public PRT (or robocars) would give them. They can't imagine a public
vehicle being as convenient (or even more convenient) than their
private automobile because they have no experience with it. They
*equate* private ownership with freedom when it comes to
transportation, so they are reluctant to give it up.

But in other domains where private ownership provides no major benefit
over the publicly-leased model, the need for "ownership" fades, as
with entertainment and the Internet.

Mike C

[eph]

Why do people buy houses when they can simply rent them? Why would
the reasons be different for cars?

F.

[Mike C]

On Sep 8, 7:48 pm, Brad Templeton <brad...@gmail.com> wrote:
> Well, the strange thing was, I seemed to be getting people here saying
> that even with robocars (which, as I define them, have no need for a
> guideway) it would be better to build guideways than run the robocars
> on existing streets.  That's the thing I can't figure.

But wouldn't it be better to get thru traffic off our streets so we
can actually walk on those streets? The elevated guideway frees the
streets of disruptive and dangerous vehicles. You can't deny that (all
other things being equal) an exclusive guideway system would be safer
than a mixed-guideway system, and we already know what vehicle traffic
does to city streets, so why *not* elevate the traffic?

That's what I can't figure: why the idea of vehicles running on
streets seems to be so attractive that we not only accept it as the
norm, but we are downright reluctant to give it up. Automobile traffic
(robotic or otherwise) on streets is A Bad Thing. It's a necessary
evil in today's society. It's something we should take pains to
eliminate, not preserve. That's the idea of lightweight elevated
vehicles: replace an enormous evil (cars on the street) with a MUCH
smaller evil (a tiny slice of visual intrusion).

Mike C

[Mike C]

I'm not saying people wouldn't still own cars - they would, probably
for a long time. But they wouldn't *require* ownership in a transit
system once they realize that equivalent freedom can be provided
without ownership. They may own one car instead of two and use the PRT
to commute daily. Or they may own as many cars as they own today, but
use them less frequently, especially in cases where the PRT is more
convenient (i.e. to go places with limited parking).

Remember, most systems don't eliminate the automobile, so nothing is
lost anyways. But for the (IMO) majority of people for whom car
ownership is about freedom of movement as opposed to status, there
will be a trend to less automobiles ownership and more use of public
facilities, provided those facilities give the same level of freedom.

Also, homes are very different than cars. People buy homes for many
reasons (equity, security, stability) that don't necessarily apply to
cars. And even then, many still choose to rent, and many others buy
condos where there is certainly not the same level of "private
ownership" as you'd get in a single family home.

People own cars because they must - there is no other model that gives
them the level of freedom and convenience they require. The exception
is in dense transit-oriented cities, where transit is so developed and
auto use so impeded that even traditional transit can exceed the
freedoms of the automobile - and there we find millions living happily
without private ownership of their transit.

Mike C

> > Mike C- Hide quoted text -

[Walter Brewer]

Suggestion: Sell PRT/Robocar as your "always available personal chauffer"

Walt Brewer

----- Original Message -----
From: "Mike C" <mwil...@gmail.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Tuesday, September 09, 2008 9:37 AM
Subject: [t-i] Re: PRT obsoleted by robocars

[gary]

I own mine because it's an inventment and a tax write-off. I don't
think either of these would apply to PRT vehicles.

gary

[gary]

Mike,

I walk on my streets all the time. And ride my bike. For the most
part this is totally not a problem. There are a few tight spots where
I wish the vehicles were computer controlled and on rails, so less
likely to run me over, but other than that I have absoltuely no
problem with ground transportation serving ground destinations.

gary

[eph]

Well, when someone rents something, they make a profit. If you own
it, you don't pay that profit. In Canada, your house isn't a tax
write-off. But you can "pimp your ride" if you own it!

You know, looking at your own situation for a solution for everyone
may not be the best strategy. If you look at rent vs own statistics I
think you will see a majority like to own stuff.

F.

[Bruce Attah]

I dislike walking alongside busy streets, but I enjoy walking where
motor traffic is absent. Cycling, too. I think I am not unusual in
this. The hobby of country walking is very popular, but don't think
I've ever met anyone who's hobby was to walk up and down arterial
roads.

[Bruce Attah]

Home ownership varies a lot according to the legal and policy climate
(tax, planning laws, security of tenure, etc.) and historical factors.
The majority of Germans tend to ask, "why own when you can rent?".

[Mike C]

Yes, of course we walk and ride all the time, but wouldn't it be nice
if we could do so without having to compete with 6+ lanes of thru-
traffic? You yourself said you prefer clear streets:

http://groups.google.com/group/transport-innovators/browse_thread/thread/6a17e049bd3296ce/6a3eedcc2a1cec26?lnk=st&q=#6a3eedcc2a1cec26

yet you object to proposals which would potentially do precisely that.

Multi-lane vehicle traffic at street level *does* have an adverse
effect on neigborhoods, and that will not change with automation. So I
can't understand why there is such objection to systems that would
tend to move vehicles out of the way and reclaim the streets for
humans. If the trip from A to B passes through C, why must the
neighborhood around C be impacted when it's just as easy for those
vehicles to pass by overhead?

Mike C.

[Bruce Attah]

Brad Templeton wrote:

> Sure, you can come up with all sorts of advantages for public
> ownership, or rather in this case, private fleet ownership. I
> understand them all, but they are all dwarfed, *dwarfed* by the big
> advantage of private ownership. Innovation. A bustling market of
> entrepreneurs dreaming up products to sell to individuals, with vast
> choice and competition, and people buying expensive new things because
> they are cool -- that's where innovation comes from. It's not
> impossible in other situations, but this is where it really thrives.

Competition of vehicle and system manufacturers to supply robotaxi
service operators who are in turn competing to provide the most
attractive service to passengers will provide an excellent impetus to
innovation -- arguably, a better set of incentives than private
ownership, because the taxi service operators will be aiming to
provide a great service as a whole (systems of communication,
scheduling, maintenance, traffic management, etc.), whereas private
owners will be focused almost exclusively on the physical car.

The other green and economic advantages already mentioned of reduced
private ownership of cars, such as fewer cars built and less land used
for parking, are far from trivial. As a bonus, this is one of those
situations where being green saves money for the consumer.

[gary]

It's definitely a pleasant experience walking through our mainstreet
when it's blocked off for our farmer's market, as I did indeed
mention, but that's more of a social thing as well...all the neighbors
there getting reacquainted, etc. But when I walk (or ride my bike) to
various destinations about town, that's a different experience. A bit
of exercise, fresh air, and an opportunity to take in the sights.
That's the situation I was describing above as far as not minding
mixing with automobile traffic.

The problem with the elevated thing that NONE of you has adquately
addressed is the drastic lack of flexibility with regards to stops.
With the automobile (or earth-based PRT) I can (and do) easily stop in
front of hundreds of places within a mile or two of my house. With an
elevated system you can't possibly have that many elevators. If you
do, the idea of a non-obtrusive elevated structure (which I never
bought into anyway) instead becomes a tangle of door-to-door
elevators. The vision many of you share of an elevated PRT system is
one with far fewer destinations than a car, closer really to a
lightrail design. And I never take lightrail.

gary

On Sep 9, 8:08 am, Mike C <mwill...@gmail.com> wrote:
> Yes, of course we walk and ride all the time, but wouldn't it be nice
> if we could do so without having to compete with 6+ lanes of thru-
> traffic? You yourself said you prefer clear streets:
>

> http://groups.google.com/group/transport-innovators/browse_thread/thr...

> > - Show quoted text -- Hide quoted text -

[Walter Brewer]

While the public system as a whole is a monopoly, can't there be private
operators using the guideways in competition. and seeking ridership and
profit?
Some public transit systems now use competitieve bidding for bus and
LRToperations, although not in competition on same routes that I know of.

Walt Brewer
----- Original Message -----
From: "Bruce Attah" <bruce...@googlemail.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Tuesday, September 09, 2008 11:11 AM
Subject: [t-i] Re: PRT obsoleted by robocars

>
>
>

[Mike C]

Gary,

Elevators are not required for all stations. Some systems (Mister)
have all stations at the ground level and provide steep guideway
climbing to facilitate it. Even without steep climbing, half-height
stations (vehicles and passengers meet at 8-feet height, with 10%
grade ramps for both vehicles and passengers) could be used to avoid
the elevator. And, of course, putting PRT right into buildings is the
ideal, because they already have elevators.

More to the point, elevated stations with elevators are NOT a
requirement of PRT - that's just the default implemementation when
other options are not available. Steep climbing PRT can avoid
elevators completely by putting all stations at grade. Steep climbing
is perhaps the best argument for suspended designs, because supported
designs are generally limited by the angle of the cabin.

Mike

[gary]

Mike,

It's always a bit difficult to discuss the elevation idea because as
you point out, there are several different flavors of it represented
by the people in this group. As for the steep climbing alternative,
wouldn't that create even more of an obstruction than elevators?

Maybe it would help if we had a new column in the spreadsheet...the
minimum distance between stations. For example if two destinations
are only 25 feet apart, presumably NONE of the proposals would treat
that as two stations. But at 200 feet apart, presumably some would
and some wouldn't. That would certainly help me understand how people
here are perceiving elevation.

gary

[Dennis Manning]

Mike:

Good points. It's simply that PRT can facilitate much different land use
patterns than at grade robocars.

Dennis

----- Original Message -----
From: "Mike C" <mwil...@gmail.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Tuesday, September 09, 2008 9:31 AM
Subject: [t-i] Re: PRT obsoleted by robocars

[eph]

True, but to put things into perspective, even in Germany where they
have to lowest home ownership rates in Europe, it's at 40% (and
rising).

F.

[David Maymudes]

Here's my problem, Gary:

Assumptions:
1) I want to build a system in American cities in the next 20 years.
2) I think it's not possible in the next 20 years to run automated vehicles
together with cars.
3) I think it's not possible in the next 20 years to get rid of all manually
driven surface vehicles.

This leads me to conclude that I need a grade-separated system. Typically,
at least in the parts of Seattle I care about, tunneling is too expensive
and land isn't available for new at-grade right of way. This leads me to
conclude that elevated is the only option.

Which piece of that argument do you disagree with? I agree that it would be
nice to have something that stopped at every house and was really
inexpensive, but I'm not willing to give up on Assumption #1 above.

--David

-----Original Message-----
From: transport-...@googlegroups.com
[mailto:transport-...@googlegroups.com] On Behalf Of gary
Sent: Tuesday, September 09, 2008 11:27 AM
To: transport-innovators
Subject: [t-i] elevation

Brad,

I know we differ on issues like rails, ownership and such, but I am
completely at a loss to understand why everyone in this group is so
hell bent on elevating their systems. I see it as absolute insanity.
Too ugly, too inflexible, and the real show-stopper, impossibly
expensive. So if you're feeling this way lately, don't feel like
you're the only one!

gary

[Dennis Manning]

The Germans must be better landlords :-)

----- Original Message -----
From: "eph" <rhaps...@yahoo.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Tuesday, September 09, 2008 11:35 AM
Subject: [t-i] Re: PRT obsoleted by robocars

>

> True, but to put things into perspective, even in Germany where they
> have to lowest home ownership rates in Europe, it's at 40% (and
> rising).
>
> F.

snip

[Brad Templeton]

Vehicles on streets are more attractive than guideways for a whole
bunch of reasons that I have outlined. We already have the streets,
and building the guideways is expensive, and in the opinion of many,
view-blocking. But the view-blocking is a minor thing for me. Even
the expense may be secondary to a much bigger issue, which is the
flexibility of roads. Roads don't lock you in to a particular
vehicle width or drive choice. People can build almost any sort of
vehicle they imagine to go on roads -- 2 wheels, 3, 4, 6. Even one
wheel. Even a hovercraft if you could make it stable. Even a
walking robot like bigdog. A guideway or rails locks you into that
choice for many decades. Once the guideway is in place you can't go
on it unless you fit it, and don't go over its weight limit. And
follow the guideway owner's rules (probably the city.)

Don't you have to follow rules of the road? You do, but these are an
entirely different class of rules. And in fact, in a robocar world
you really only need two rules -- do not endanger others, and do not
impede others unfairly. Those rules apply everywhere in life. A
guideway has a ton of extra rules.

Why are rules so bad? Because the real revolution, the real
innovation, will be something none of us have yet thought of. I
don't know what that innovation is, but I know how to frame the world
to make it more likely, and that's with the most general system I can
think of, which is generic paved ground.

This debate goes on everywhere in technology. There are people who
want the "network" to be smart, and people who want the network to be
a dumb platform, and let the innovation take place in what goes over
it. In the digital world, we call the first type of thinking
bellhead, old-world thinking, and the latter type is nethead, internet-
style thinking. When IP was designed, none of the people who designed
it knew all the amazing things that would be done. They just knew if
they made it simple and flexible and got out of the way, great things
would happen.

All the time there are people who come up with great ideas and
advantages of what a centralized system could do, and why it is
clearly better than a simple system. The arguments always seem
convincing, because they are comparing some great new idea with an
idea that has yet to be thought of. And yet, they are almost always
totally wrong.

So this temptation must be avoided. Build the simplest possible
platform and let the people who put things on the platform do the
innovating. Get out of their way.

[Frank Randak]

CBS News Productions in New York is producing a 14 part series on new ideas
and technology called NextWorld, which is being shown on Wednesdays at 5 and
9 PM. In July, they were looking on the Internet for new ideas in
transportation and they found the AVT. After reviewing our revised web side,
www.Solatrek.com, they concluded that the AVT is "the only viable solution"
to our freeway traffic problems.

On July 17th, CBS News Productions spent all day at California Lutheran
University (CLU) filming and interviewing the AVT Project. We don't know yet
the date for the AVT segment, but we do have written confirmation from CBS
News Productions that the AVT will be included in the NextWorld series.

Last month the AVT Project was interviewed on Ventura County's KKZZ talk
radio for a full hour and we have been invited to make another presentation
at the KKZZ sponsored Big E Extravaganza (E for Environment) on September
27th and 28th in Oxnard, with the 16 foot AVT prototype model.

We will be presenting "The AVT - a Solar Powered Energy and Traffic
Solution" at CLU on September 24th at 7:30 PM. If you are able to attend,
please call 800-924-3306 to RSVP and get directions.

Thank you,
Frank Randak

[Brad Templeton]

Yes, though it's not as good. But this is a minor point. I avoid
prognostication on how percentage of vehicles will be in fleets and
which will be privately owned because that's a choice that will be up
to the people and the market, and it would be foolish to try to force
that choice. It will be what it is, and it doesn't make a great deal
of difference to the results. I personally expect lots of both. I
also expect (and this is more important) private tinkering to be a
source of innovation. It should be possible to do that safely for a
few reasons. One, certifying safety will be one of the governmental
roles here, in all probability. Two, it will be possible to buy pre-
certified safety systems which watch over an untrusted experimental
control system and block it from doing unsafe things. (The operator
of an experimental car faces the risk that sometimes their car may
switch to fail-safe mode and they will lose the cool new experimental
features.)


But again, this is a minor point. It will be a question of tastes,
and predicting tastes is fun, but largely fruitless.

[Dennis Manning]

Brad:

You make some very interesting observations. I'm not in your league when it
comes to the internet and computers, but I have read a lot of George Gilder
and believe I have a reasonable grasp of the "smart network" vs. "dumb
network" ideas.

So far as I know there are quite a lot of each out there with most of the
explosion occurring on the edge, but what I've been saying is elevated PRT
can yield land uses and patterns not possible with wheeled road cars. Think
what elevators did for building heights. (no visual problem there) :-)

The unknown innovation yet to come is not confined to the transportation
portion only. You should know that better than anyone.

Dennis

----- Original Message -----
From: "Brad Templeton" <bra...@gmail.com>
To: "transport-innovators" <transport-...@googlegroups.com>
Sent: Tuesday, September 09, 2008 11:57 AM
Subject: [t-i] Re: PRT obsoleted by robocars

[Mike C]

Gary,

Responses below.

On Sep 9, 1:21 pm, gary <garydst...@gmail.com> wrote:
> Mike,
>
> It's always a bit difficult to discuss the elevation idea because as
> you point out, there are several different flavors of it represented
> by the people in this group.  As for the steep climbing alternative,
> wouldn't that create even more of an obstruction than elevators?
>

How so? If a PRT system climbs at 45-degrees, that means it needs 1
foot of clearance for every 1 foot of descent space. So, assuming 12-
ft high vertical clearance is required, that means that the descent to
ground level would require a ground area of 12ft long and about 5 ft
wide (approximate width of the vehicle). This smaller than a typical
parallel parking space. Double it for the ascent track, and that's two
parallel parking spaces consumed at ground level for the lead-in
tracks. Add 20 linear feet for a two berth station, and you can move a
few hundred people per hour in the space of 4 parallel parking spaces,
with no elevator required.

Speaking of parking, how many parking garages are there in a typical
city? Each one of those garages can integrate a PRT station on the
second floor, and the elevator comes for free.

> Maybe it would help if we had a new column in the spreadsheet...the
> minimum distance between stations.  For example if two destinations
> are only 25 feet apart, presumably NONE of the proposals would treat
> that as two stations.  But at 200 feet apart, presumably some would
> and some wouldn't.  That would certainly help me understand how people
> here are perceiving elevation.
>

But the point is, PRT station density is whatever you want it to be.
Some PRT systems do it better than others, but all PRT designs I've
seen assume a large number of lightweight stations scattered densely
throughout the city. Maybe it's not the theoretical infinite station
density of an actual taxicab, which can drop you off literally
wherever the street takes you, but they can be made close enough for
most people to accept it. And if PRT goes into buildings themselves,
many times it will bring you even closer than road-based vehicles can
do.

Mike C.

[Brad Templeton]

Sure, interesting things could happen if cars got off the roads and
into the sky or underground.

But if you think I'm way too optimistic in hoping for safe robocars in
15 years, I have to say the dream of a complete elevated
transportation network goes far beyond that. Not merely pie in the
sky, but a complete guideway system up there.

While both problems are hard, I will bet on the robocars because they
can happen one at a time. Once demonstrated safe, it's worthwhile for
a single person to buy a single robocar. There is no cliff of
adoption to go over. Convincing the world to shut down car traffic on
a network of streets, or a whole city, seems beyond doable, not for
technical reasons, but political ones.

If we really need to go into the sky, I suspect it would be more with
a tire style guideway as ULTra has, so that it's still the same
platform. Though again, the guideways would dictate what went on them
in a way that roads don't.

There's a lot you can do at-grade, by the way, if you want to summon
up the same political will needed to replace the roads.

For example you could convert some streets only to much smaller
vehicles, or segway-style vehicles except for the disabled, if you
want to turn streets into parks.

Another thing about elevated guideway -- I saw somebody here talk
about 12' high guideway, which is not enough for a lot of vehicles on
the road today. What if somebody wants to be able to drive the
Purple Palace (http://www.purplepalace.org/) down the road? You
may or may not say that's worth it, but who are you to say? (Yes, I
know that overhead wires already block this, as do highway overpasses,
which is why the purple palace has to be stored within 50 miles of the
playa.)
On Sep 9, 12:25 pm, "Dennis Manning" <john.manni...@comcast.net>
wrote: