Driverless Car ?!
Jean
What are the major research work happening in this area ?
regards
Jean
zzbu...@netscape.net
> When can I buy one ?
> What are the major research work happening in this area ?
The research is so active after 100 years of automotive idiots
running the world,
it's already moved full swing into drones, pilotless airplanes
ships, and submarines,
And laser-guided lasers, OnLine Publlishing, neo Wind Energy, Post
McDonald's
Holograms, post Ford Batteries, and Post GM Robotics.
>
> regards
> Jean
Androcles
"Jean" <aler...@rediffmail.com> wrote in message
news:a9b4a58e-f293-4b52...@41g2000yqf.googlegroups.com...
> When can I buy one ?
Adapt your own. All it really needs is a computer.
> What are the major research work happening in this area ?
Artificial intelligence.
Any car with cruise control only has to be steered. The easiest
way to overcome that problem was solved in 1829 - it was called
"rail".
http://en.wikipedia.org/wiki/Stephenson's_Rocket
Today, any industrial vision system with edge detection
can find the side of the road, so all you need is a servo on the
steering.
http://en.wikipedia.org/wiki/Edge_detection
Add forward looking radar to slow behind the vehicle
in front, many cars already have anti-collision radar behind,
http://www.guru3d.com/newsitem.php?id=4304
and there isn't much else you need do except park. That's
where the AI comes in.
Ian Parker
> When can I buy one ?
> What are the major research work happening in this area ?
>
provided so far. The "gold standard" for urban driving is the DARPA
challenge.
http://www.theregister.co.uk/2007/11/04/cmu_wins_urban_challenge/
On motorways and rural roads driving can be done in a more controlled
environment, you can drive at 130km/h nose to tail as I think the
other contributors have implied.
As far as "when will they be on sale?" To some extent that depends on
legal/political considerations as well as technical. At the present
state of technology you can have automatic freeway lanes, their
introduction would involve a political decision. One thing is clear, a
road should if possible be completly automatic, automatic and non
automatic driving should not mix. This means that after a set date all
cars will have tio be equiped with a certified system.
There are legal questions about the libability for accidents. These
would generally be fewer than now, but there will still be some
failures. A legal framework needs to be developed.
There is one possibility which is quite interesting. Aircraft have
transponders. Could cars driving on automated roads also have GPS
transponders which could be used for automatic scheduling.
One of the first problems to hit Obama's desk will be Detroit and how
to bail it out. Might I suggest that the development of automation
might be a part of the answer. The industry desparately needs
restructuring.
Things like "a railway" do not help. The whole point of a car is that
you can not only drive from Washington to New York, you can drive
along you neighbourhood road up to your front door. Public transport
cannot provide this. You can also drive along a route where public
transport would involve several changes and waits.
- Ian Parker
ji...@specsol.spam.sux.com
> On 15 Dec, 08:45, Jean <alertj...@rediffmail.com> wrote:
>> When can I buy one ?
>> What are the major research work happening in this area ?
>>
> I think you deserve a slightly better answer than what has been
> provided so far. The "gold standard" for urban driving is the DARPA
> challenge.
>
> http://www.theregister.co.uk/2007/11/04/cmu_wins_urban_challenge/
>
> On motorways and rural roads driving can be done in a more controlled
> environment, you can drive at 130km/h nose to tail as I think the
> other contributors have implied.
>
> As far as "when will they be on sale?" To some extent that depends on
> legal/political considerations as well as technical. At the present
> state of technology you can have automatic freeway lanes, their
> introduction would involve a political decision. One thing is clear, a
> road should if possible be completly automatic, automatic and non
> automatic driving should not mix. This means that after a set date all
> cars will have tio be equiped with a certified system.
Not "should", "must"; mixing is a receipe for disaster.
> There are legal questions about the libability for accidents. These
> would generally be fewer than now, but there will still be some
> failures. A legal framework needs to be developed.
The legal framework is already there.
When the system fails, and it eventually will as all things do, and
a SUV full of soccer moms and small children gets wiped out, the deep
pockets systems vendor gets sued out of existance.
> There is one possibility which is quite interesting. Aircraft have
> transponders. Could cars driving on automated roads also have GPS
> transponders which could be used for automatic scheduling.
GPS doesn't have anywhere near the accuracy required to keep a vehicle
centered in a lane; it takes cables in the roadway.
The sheer number of vehicles means the bandwidth required to talk to
them all becomes a wishfull fantasy.
> One of the first problems to hit Obama's desk will be Detroit and how
> to bail it out. Might I suggest that the development of automation
> might be a part of the answer. The industry desparately needs
> restructuring.
Automation that eliminates the current union contracts would help.
--
Jim Pennino
Remove .spam.sux to reply.
tj Frazir
mostly weekends
Coater
"tj Frazir" <Gravity...@webtv.net> wrote in message
news:13289-494...@storefull-3213.bay.webtv.net...
> shit they are evrywhere ,
> mostly weekends
>
crap! I caught myself agreeing with tj
zzbu...@netscape.net
> On 15 Dec, 08:45, Jean <alertj...@rediffmail.com> wrote:> When can I buy one ?
> > What are the major research work happening in this area ?
>
> I think you deserve a slightly better answer than what has been
> provided so far. The "gold standard" for urban driving is the DARPA
> challenge.
>
> http://www.theregister.co.uk/2007/11/04/cmu_wins_urban_challenge/
>
> On motorways and rural roads driving can be done in a more controlled
> environment, you can drive at 130km/h nose to tail as I think the
> other contributors have implied.
>
> As far as "when will they be on sale?" To some extent that depends on
> legal/political considerations as well as technical. At the present
> state of technology you can have automatic freeway lanes, their
> introduction would involve a political decision. One thing is clear, a
> road should if possible be completly automatic, automatic and non
> automatic driving should not mix. This means that after a set date all
> cars will have tio be equiped with a certified system.
>
> There are legal questions about the libability for accidents. These
> would generally be fewer than now, but there will still be some
> failures. A legal framework needs to be developed.
>
> There is one possibility which is quite interesting. Aircraft have
> transponders. Could cars driving on automated roads also have GPS
> transponders which could be used for automatic scheduling.
Well, you don't have to worry about it. Civilian GPS is so
overblown,
in regards to anything and everything, pretty soon it will come
free with
every movie ticket you buy.
Bernd Felsche
>Ian Parker <ianpa...@gmail.com> wrote:
>> As far as "when will they be on sale?" To some extent that
>> depends on legal/political considerations as well as technical.
>> At the present state of technology you can have automatic freeway
>> lanes, their introduction would involve a political decision. One
>> thing is clear, a road should if possible be completly automatic,
>> automatic and non automatic driving should not mix. This means
>> that after a set date all cars will have tio be equiped with a
>> certified system.
>Not "should", "must"; mixing is a receipe for disaster.
>> There are legal questions about the libability for accidents. These
>> would generally be fewer than now, but there will still be some
>> failures. A legal framework needs to be developed.
>The legal framework is already there.
Assuming that there would be fewer crashes with automated cars on
public roads is based on what? That the automated systems will make
fewer mistakes?
>When the system fails, and it eventually will as all things do, and
>a SUV full of soccer moms and small children gets wiped out, the deep
>pockets systems vendor gets sued out of existance.
The components of the system don't even have to fail. The system
simply has to encounter a situation that the designers didn't
foresee.
Like when the automated storm blows automated debris across the
automated road in front of the automated cars.
Driving a car in traffic is an extremely complex task. It takes a
human months to acquire some basic skills and years to acquire the
experience and judgement to be reasonably safe.
Automated systems have not even one percent of the cognitive
abilities of a good driver. Situational awareness is part of what
makes good drivers. It allows them to anticipate and to avoid risky
situations. The responsiveness of automated systems cannot make up
for that, mainly because of the physical inertia of the vehicle and
the limits of adhesion and cornering.
We have, in the past, automated systems to ease the burden, the
risks of the workplace and the tedium of human existence. So why
automate driving? It's not that risky. (It's more dangerous to be
admitted to hospital for treatment.) It's hardly a burden with
power steering, etc. And it's only tedious if you allow it to be.
Driving can in fact be quite enjoyable and fulfilling.
To those who think that they can automate driving; you can't:
Technologically speaking; there's still a gap the size of the 20th
century. If the technologists/dreamers can't see that, then they
can't drive.
>> There is one possibility which is quite interesting. Aircraft have
>> transponders. Could cars driving on automated roads also have GPS
>> transponders which could be used for automatic scheduling.
>GPS doesn't have anywhere near the accuracy required to keep a vehicle
>centered in a lane; it takes cables in the roadway.
Back in the 1920's and 1930's, Germany began building its Autobahn
network. First along the lines of railways, with very long
straights, deep cuttings, etc. But the Engineers driving along those
new stretches recognized a problem straight away; it wasn't
involving the driver any more. Since then, Autobahns have had
meandering, sweeping bends that follow the contours of the
countryside, using only deep cuttings, tunnels and bridges where
gradients are actually too steep for cars and trucks; or where going
around the mountain/valley would add too much distance to the
journey.
Cars are not trains. Roads and railroads.
The "scheduling" of cars is a nonsense. If merging is what is meant,
then that has to be done in the immediate proximity of the vehicles
concerned; not on the other side of a continent at 100 milliseconds
latency.
Where manual merging doesn't work, it's only because of the
attitudes of the majority of drivers.
>> One of the first problems to hit Obama's desk will be Detroit and
>> how to bail it out. Might I suggest that the development of
>> automation might be a part of the answer. The industry
>> desparately needs restructuring.
>Automation that eliminates the current union contracts would help.
The easiest path to that is the one with the highest step at the
beginning: Not bailing out the car makers.
Those car makers who want to survive, will find a way to survive.
Those who don't, or can't because of external encumberances, will
close. Greed doesn't begin in the board room.
--
/"\ Bernd Felsche - Innovative Reckoning, Perth, Western Australia
\ / ASCII ribbon campaign | Second to agriculture, humbug is the
X against HTML mail | biggest industry of our age.
/ \ and postings | -- Alfred Nobel
jmfbahciv
<snip>
>
>> One of the first problems to hit Obama's desk will be Detroit and how
>> to bail it out. Might I suggest that the development of automation
>> might be a part of the answer. The industry desparately needs
>> restructuring.
>
> Automation that eliminates the current union contracts would help.
It will never happen. I figure at least two will have to be dissolved
into bankruptcy before the unions change their mindset. There
are two sets of managements that require mindset changes: the
coroporate management and, more importantly, the union management.
/BAH
Yevgen Barsukov
wrote:
> j...@specsol.spam.sux.com wrote:
> >Ian Parker <ianpark...@gmail.com> wrote:
> >> As far as "when will they be on sale?" To some extent that
> >> depends on legal/political considerations as well as technical.
> >> At the present state of technology you can have automatic freeway
> >> lanes, their introduction would involve a political decision. One
> >> thing is clear, a road should if possible be completly automatic,
> >> automatic and non automatic driving should not mix. This means
> >> that after a set date all cars will have tio be equiped with a
> >> certified system.
> >Not "should", "must"; mixing is a receipe for disaster.
> >> There are legal questions about the libability for accidents. These
> >> would generally be fewer than now, but there will still be some
> >> failures. A legal framework needs to be developed.
> >The legal framework is already there.
>
> Assuming that there would be fewer crashes with automated cars on
> public roads is based on what? That the automated systems will make
> fewer mistakes?
They are unable to willfully break the rules.
They can't get distracted by their wifes.
They can't be "deep in thought".
They don't get drunk.
They don't speed.
But all of that is not the main point.
Main point is - they have much better tools,
which can be cheaply improved even more.
We only have two stereoscopic cameras to guide
us, and best driver in the world still has only
2. All our distance and speed estimates are just that
- estimates based on prior knowledge how large objects
normally are, and extrapolations about where they are
right now and how they might continue their motion when
we are not looking.
Good robotic drivers (once that won or succesfuly completed off-
road and urban DARPA challenge) have multiple range lasers and radars
to their disposal that give them not an estimate but an exact
MEASUREMENT of the distance and speed of all objects around them. They
never
have to extrapolate anything - they can read with millisecond sampling
rate all distances and speeds at all times.
Of cause all that takes significant processing power, both
parallel and serial. But in serial processing power even
a handheld phone with its 1GHz processor beats
humans with their 10 Hz neuron firing rate by factor of
10^8. This is what affects reaction time, and that is
critical for safe driving.
In parallel processing humans still have an edge, but
barely. And considering all the specialized measurement equipment,
that much parallel processing is
not even needed, because most of the 3D navigation tasks
driver performs are needed to replace that ranging equipment he does
not have.
So to summarize - robotic driver with laser and radar ranging
equipment + cameras and adequate software is already superior to human
driver from hardware standpoint
DARPA challenges demonstrated that it can be done not
even one way, but many different ways. It is not even
very expensive.
>
> >When the system fails, and it eventually will as all things do, and
> >a SUV full of soccer moms and small children gets wiped out, the deep
> >pockets systems vendor gets sued out of existance.
>
> The components of the system don't even have to fail. The system
> simply has to encounter a situation that the designers didn't
> foresee.
>
> Like when the automated storm blows automated debris across the
> automated road in front of the automated cars.
A lot of that has been tested in DARPA challenge.
If experiencing some unexpected situation, there will be
a safe mode. Car will just drive to the curb and stop, while waiting
for situation to come back to one that it can
handle. Nothing futuristic - active cruise control is luxury cars is
already able to do the same thing.
Btw. that is a good strategy to human driver as
well.
That is not to say that software with a lot of hevristics
is not important. It is. Human drivers have these hevristics too:
"look over your shoulder before changing lanes", "look behind you when
you are backing", "slow down on icy road", "look at the right side of
the road if blinded by lights of approaching vehicle" etc.
The list of hevristics will keep improving. But human drivers have a
choice to use or not to use some generally known good practices, while
automatic software will not have such choice, it will just keep using
all the good stuff that is being added.
>
> Driving a car in traffic is an extremely complex task. It takes a
> human months to acquire some basic skills and years to acquire the
> experience and judgement to be reasonably safe.
>
> Automated systems have not even one percent of the cognitive
> abilities of a good driver. Situational awareness is part of what
> makes good drivers. It allows them to anticipate and to avoid risky
> situations. The responsiveness of automated systems cannot make up
> for that, mainly because of the physical inertia of the vehicle and
> the limits of adhesion and cornering.
All of these limits are simple 5th grade physics.
It can be all put into the machine, and is already
done with electronic stability control and ABS breaks.
The very fact that these devices already exist and statistics show
that they reduce accidents is a proof
that most human drivers are not good at this predictive
physics (regardless if they do it consciously or intuitively) and can
be successfully replaced with simple but well informed and diligent
calculator.
>
> We have, in the past, automated systems to ease the burden, the
> risks of the workplace and the tedium of human existence. So why
> automate driving? It's not that risky.
Death in a car accident is the most common cause
of death in USA and most other countries.
>(It's more dangerous to be
> admitted to hospital for treatment.) It's hardly a burden with
> power steering, etc. And it's only tedious if you allow it to be.
>
> Driving can in fact be quite enjoyable and fulfilling.
Yes, but most of the time it is a waste of your cognitive
abilities that could be better spent thinking about something useful
instead of intuitively making mind-bogglingly complex 3D estimates of
"how fast is this little dot on the opposite lane is approaching".
Mind you "intuitively" means that a big share of your 10^10 brain
neurons (parallel processors) are actually working to do it. It is not
"free" in any way. It is taking a lot of your bandwidth away from
other thoughts.
Maybe you will make a wrong decision about what to tell
your wife or what to buy on stock market because you missed
that little bit of extra bandwidth.
You can still drive for sports just like you go fishing
for sports. But I would not like to rely on my fishing
abilities to put food on the table every day.
>
> To those who think that they can automate driving; you can't:
> Technologically speaking; there's still a gap the size of the 20th
> century. If the technologists/dreamers can't see that, then they
> can't drive.
See DARPA challenges results. After first DARPA off-road
challenge in 2004 which none of the cars completed, everyone was
saying what you are saying. But that was 4 years ago. Just a year
later, in 2005, 5 cars completed the challenge:
http://en.wikipedia.org/wiki/DARPA_Grand_Challenge#2005_Grand_Challenge
More recently, in 2007, 6 cars have completed Urban
driving challenge:
http://en.wikipedia.org/wiki/DARPA_Grand_Challenge#2007_Urban_Challenge
The progress will not be stopped (that is, as long
as there is still oil left to fuel it).
Regards,
Yevgen
tj Frazir
Benj
> in 50 years there wount be any cars.'
You got it! Nothing but "light rail" in a communist utopia!
A nice peaceful return to the 1900's!
Those were the days!
We are so "Green"!
tj Frazir
the second dark age has begun.
endless groath on earth is imposible.
35 billion acres of land with just 6 billion farmable acres.
the top 1 % of the population will stangle the planet over monopoly
controle.
the 38,000 multimilionaires in usa will nuke he planet befor 10 years
thinking 20 % and them selves in the south pacific will own a new
planet.
zzbu...@netscape.net
Well, the only thing idiot multimillionaires have EVER owned is junk
in the South Pacific.
Which is still why the people with actual science-engeeering-
aviation-economics brains still
just build satellites, GPS, Drones, AAVs, AUVs Cruise Missiles,
Phalanx, A,I, Parallel Processing,
RISC computers, Microcomputers, Fiber Optics, E-Publishing, On-Line
Publishing, USB, XML,
Neo Wind Energy, Neo Solar Energy, Neo Batteries, Pv Cells, Robots,
Biodiesel.
lasers, masers, Microwave ovens, CD, DVD, HDTV, and Post
McDonald's Holograms.
zzbu...@netscape.net
wrote:
> j...@specsol.spam.sux.com wrote:
Well, no understands beyond the number 9 in science, engineering,
technology,
economics, and automation does anything with automotive porkers
anyway,
That's why nuclear energy, neo solar energy, neo wind energy, Pv
Cell energy,
Adaptive AI, Fiber Optics, Anti-Spam, non moron code compilers,
Optical Computers,
Microcomputers, RISC processors, parallel processors. USB, XML,
laser-guided lasers.
On-Line Banking, E-Publishing, On-Line Publishing, CD+rw, DVD-rw,
DVD-rom DVD-ram, HDTV, Cruise Missiles, Drones, AAVs, AUVs,
Digital-Terrain Mapping, Biodiesel, Post McDonald's Holograms, and
GPS
come up so often in discussions with the stooges.
tj Frazir
i never neaded 1 billion ya cant stop and put evryone out of work ya
know.
hhc...@yahoo.com
Jean, the US Government funds a research project known as IVHS (or
IVHP..I forget which) under the DOT. Yes, this probject includes self
operated cars and trucks, and roadways that supply traction energy and
navigational guidance to the vehicles.
Unfortunately, this is only funded research. If the investment is
rewarded, the results will likely not be visible for another 50 or so
years,
Harry C.
jmfbahciv
> On Dec 15, 6:12 pm, Bernd Felsche <ber...@innovative.iinet.net.au>
> wrote:
>> j...@specsol.spam.sux.com wrote:
>>> Ian Parker <ianpark...@gmail.com> wrote:
>>>> As far as "when will they be on sale?" To some extent that
>>>> depends on legal/political considerations as well as technical.
>>>> At the present state of technology you can have automatic freeway
>>>> lanes, their introduction would involve a political decision. One
>>>> thing is clear, a road should if possible be completly automatic,
>>>> automatic and non automatic driving should not mix. This means
>>>> that after a set date all cars will have tio be equiped with a
>>>> certified system.
>>> Not "should", "must"; mixing is a receipe for disaster.
>>>> There are legal questions about the libability for accidents. These
>>>> would generally be fewer than now, but there will still be some
>>>> failures. A legal framework needs to be developed.
>>> The legal framework is already there.
>> Assuming that there would be fewer crashes with automated cars on
>> public roads is based on what? That the automated systems will make
>> fewer mistakes?
>
> They are unable to willfully break the rules.
<snip>
It's code and code breaks very easily because the physical
constraints change.
/BAH
Jeff▲Relf
Bernd Felsche
>It's code and code breaks very easily because the physical
>constraints change.
How does one code for e.g. the German Traffic Regulations which
begin something like:
1. Basic rules
(1) Participation in road traffic demands constant caution
and mutual regard.
(2) Every traffic participant is to conduct themselves in a
manner so that no others are damaged, endangered or, more
than demanded by circumstances, obstructed or encumbered.
:-)
Well; there's no willful breaking of rules if there is no will.
Does simply obeying the rules make the roads safer? Nyet!
Yevgen Barsukov
Physical laws involved in driving are pretty much just
Newtonian physical that have been around for over 200 years. They
don't change and can be hard-coded.
Measured parameters that are put into the Newtonian
formulas do change, but that will not break the code,
it will be used by the code.
Non-physical issues such as negotiating between several
cars about who has priority, overtaking etc have to
be programmed and this can break (and will be improved
over and over). But humans are also not
infallible in making such decisions. At least robo-car
will not drive through a 4-way stop sign because it does not feel like
stopping today. It will get into negotiation, and it is much less
likely to fail because it does not involve somebody's mood, how
insecure somebody feels this morning etc.
Regards,
Yevgen
>
> /BAH
Yevgen Barsukov
wrote:
Da! But that is not the main point. Main point is that
they will have also much better ranging tools (lasers, radars, etc)
which are always active, never look another way, never distracted.
This way robo-driver
can not only "want" to follow the rules, it also "CAN"
follow the rules much better.
Regard,
Yevgen
gabydewilde
The problem is that regulators don't know what technology is. A simple
thing to start with would be to limit the speed to the maximum speed.
It's an oxymoron, compare it to insurance companies giving discounts
if a proper alarm system is installed.
For all such technologies: If more than 5 systems are used at the same
time it can give error reports long before any critical flaw can
produce fatalities.
The Rotterdam harbor has been using trucks without drivers for more
than 40 years.
http://www.ect.nl/images/ImageArchive/HighRes/TAG080428_5952_Gepubliceerd.jpg
Humans are expensive and they make mistakes all of the time. From a
fail safe perspective there are dozens of systems that can assist
validation of machine decisions. Things like mobile phone towers, Tee
Vee and Radio stations, the position of the sun etc etc If you have
thousands of cars on the road calibrating Tee Vee navigation it will
become more accurate by the day. If it disagrees with the other 20
systems it needs to tune it self. If it (all of a sudden) disagrees
with 5 out of 20 other systems it should report it self for
maintenance. It could slide the cabin onto a new trailer at the next
parking. Fresh replacement trailers could be parked all over the
place.
Nothing could go wrong.
__________
http://blog.360.yahoo.com/factuurexpress
ji...@specsol.spam.sux.com
> Da! But that is not the main point. Main point is that
> they will have also much better ranging tools (lasers, radars, etc)
> which are always active, never look another way, never distracted.
> This way robo-driver
> can not only "want" to follow the rules, it also "CAN"
> follow the rules much better.
Absent a cable buried in the roadway, how would the car know where it is
in a lane?
GPS is not accurate enough for that.
tj Frazir
RLW
> What are the major research work happening in this area ?
Volkswagen has been doing enough research on this that you should be able
to find ample reading material by searching news aggregation sites or
wire-service sites. Even a simple Web search yields numerous articles.
Try "self-driving car".
Lofty Goat
> Absent a cable buried in the roadway, how would the car know where it
> is in a lane?
Artificial vision. Already been done, folks are now working on doing it
really well.
ji...@specsol.spam.sux.com
Sorry, wrong answer.
The only lane reference is the line down the middle of the lane, and
that can take numerous forms. Sorting out those forms is trivial for
a human and challenging for artificial vision.
Seeing the center line is difficult at best for humans at night and
in bad weather and likely impossible for a machine.
One could stripe roadways with something highly visible to both machines
and humans.
However, the problem with all these schemes is the enormous cost of
settting up and maintaining roadways such that these systems work
at all.
FWIW I've spent the last 18 years working with traffic engineers so
I have some idea what has been proposed, tried, and what works.
Yevgen Barsukov
Check out the DARPA challenge links. They have solved
this issue with laser ranging + cameras both for off-road
and urban driving.
In 2007 challenge, 6 cars designed by different independent teams all
succesfuly navigated 55 miles of Urban landscape (no cables or any
other artificial elements) with traffic lights, other cars,
pedestrians and traffic signs.
So problem of road edge detection and other traffic signs is not
only soluble but already solved in 6 different ways.
Regard,
Yevgen
ji...@specsol.spam.sux.com
Yeah, sure...
ji...@specsol.spam.sux.com
My first response was too flip...
Yeah, no artificial elements other than the roads being lined with
bright white concrete K-rail where they almost, but not quite,
reached a blazing speed of 15 MPH.
How many times did the vehicles just stop and stare trying to figure
out what to do next?
How many collisions were there in this extremely light traffic run?
None of these techno-toys would last more than 60 seconds on a real
road with real traffic.
gabydewilde
>
> None of these techno-toys would last more than 60 seconds on a real
> road with real traffic.
>
Why not? You are perfectly dumb, ignorant, unable to judge information
and we allow you on the road. Computers have memory, ability to learn,
exact math, self diagnostics. All kinds of things you don't have.
They can drive on 2 strips of road not much thicker than the tires and
do so tirelessly.
I know I know, your science is 500 years outdated...
But still, wagonways were being used in Germany as early as 1550.
I think you are ready for the concept.
>My first response was too flip...
It's only 50 years man get over yourself.
___________
http://blog.360.yahoo.com/factuurexpress
Bernd Felsche
>> Da! But that is not the main point. Main point is that they will
>> have also much better ranging tools (lasers, radars, etc) which
>> are always active, never look another way, never distracted.
And hundreds of those radar devices in close proximity aren't going
to cause a problem? They current ones are "confused" by the roadside
furniture and have difficulty picking up things like pedestrians.
>> This way robo-driver can not only "want" to follow the rules, it
>> also "CAN" follow the rules much better.
Fiction. Machines don't "want". They just do as previously
instructed. No more. No less.
>Absent a cable buried in the roadway, how would the car know where
>it is in a lane?
>GPS is not accurate enough for that.
Lane-following using the lane markings.
This isn't perfect; wet roads, snow, sand and other debris can
confuse the systems currently deployed as driving aids by BMW,
Citroën, etc.
Lane markings can further be "confusing" due to road works with
temporary markings e.g. diverting around the work site.
ji...@specsol.spam.sux.com
> ji...@specsol.spam.sux.com wrote:
>>Yevgen Barsukov <evge...@gmail.com> wrote:
>
>>> Da! But that is not the main point. Main point is that they will
>>> have also much better ranging tools (lasers, radars, etc) which
>>> are always active, never look another way, never distracted.
>
> And hundreds of those radar devices in close proximity aren't going
> to cause a problem? They current ones are "confused" by the roadside
> furniture and have difficulty picking up things like pedestrians.
>
>>> This way robo-driver can not only "want" to follow the rules, it
>>> also "CAN" follow the rules much better.
>
> Fiction. Machines don't "want". They just do as previously
> instructed. No more. No less.
>
>>Absent a cable buried in the roadway, how would the car know where
>>it is in a lane?
>
>>GPS is not accurate enough for that.
>
> Lane-following using the lane markings.
>
> This isn't perfect; wet roads, snow, sand and other debris can
> confuse the systems currently deployed as driving aids by BMW,
> Citroën, etc.
>
> Lane markings can further be "confusing" due to road works with
> temporary markings e.g. diverting around the work site.
Yeah, that's why the only thing that works reliably is buried cables.
Lofty Goat
> Lofty Goat <rlwa...@gmail.com> wrote:
>> On Wed, 17 Dec 2008 16:30:01 +0000, jimp wrote:
>>> Absent a cable buried in the roadway, how would the car know where it
>>> is in a lane?
>>
>> Artificial vision. Already been done, folks are now working on doing
>> it really well.
>
> Sorry, wrong answer.
> FWIW I've spent the last 18 years working with traffic engineers so I
> have some idea what has been proposed, tried, and what works.
Look it up on the Web. The folks doing such research post plenty of good
reading material there.
Thence you'll pick up enough of the vocabulary of the subject matter to
run journal searches. IEEE's site is a good place to start.
The application of artificial vision to self-driving cars is difficult
enough to put it at the leading edge of such research.
FWIW? So what?
Buried cable systems date back forty years. Never did work. Without
vision, cars run into things the cables can't help them avoid.
tj Frazir
when a tv dont and you dont get no warning .
spikrd magnetic cable in the center of road and radar brakes for
intertstate only .
tj Frazir
You cant have a driverless car without a driverless road
Bernd Felsche
>Bernd Felsche <ber...@innovative.iinet.net.au> wrote:
>> ji...@specsol.spam.sux.com wrote:
>>>Yevgen Barsukov <evge...@gmail.com> wrote:
>>>Absent a cable buried in the roadway, how would the car know where
>>>it is in a lane?
>>>GPS is not accurate enough for that.
>> Lane-following using the lane markings.
>> This isn't perfect; wet roads, snow, sand and other debris can
>> confuse the systems currently deployed as driving aids by BMW,
>> Citroën, etc.
>> Lane markings can further be "confusing" due to road works with
>> temporary markings e.g. diverting around the work site.
>Yeah, that's why the only thing that works reliably is buried cables.
Until there are roadworks: Car follows buried cable into roadworks
due to failed database update. :-(
ji...@specsol.spam.sux.com
You need to the your optometrist about that severe case of tunnel
vision.
Knowing where you are and avoiding things are two separate problems.
The cable serves as a reference to where the vehicle is on the roadway.
And it worked pretty well in the experiments Caltrans did a couple
of years ago.
ji...@specsol.spam.sux.com
> ji...@specsol.spam.sux.com wrote:
>>Bernd Felsche <ber...@innovative.iinet.net.au> wrote:
>>> ji...@specsol.spam.sux.com wrote:
>>>>Yevgen Barsukov <evge...@gmail.com> wrote:
>
>>>>Absent a cable buried in the roadway, how would the car know where
>>>>it is in a lane?
>
>>>>GPS is not accurate enough for that.
>
>>> Lane-following using the lane markings.
>
>>> This isn't perfect; wet roads, snow, sand and other debris can
>>> confuse the systems currently deployed as driving aids by BMW,
>>> Citroën, etc.
>
>>> Lane markings can further be "confusing" due to road works with
>>> temporary markings e.g. diverting around the work site.
>
>>Yeah, that's why the only thing that works reliably is buried cables.
>
> Until there are roadworks: Car follows buried cable into roadworks
> due to failed database update. :-(
Not an issue since the only place such things exist is on closely
monitored test areas.
The cost of such systems guarantee they will never be deployed in
the real world.
jmfbahciv
Sigh! You are not thinking at all. People are still writing
Y2K bugs. That involves dates. Do you really think that people
won't write bugs involved time? And what about the coders who
have absolutely no idea that there is a large difference between
the length of a meter and the length of a yard. Haven't you been
paying attention to what is written in these newsgroups by
people who claim to be expert coders? They don't have any
idea about dimensional analyses. They know nothing about
numerical recipes that can do calculus. Some people don't even
know basic high school algebras (there is a poster who refuses
to acknowledge distributive properties and cannot divide both
sides of an equation by 2). Furthermore, there are vehicles
which have a need to break traffic laws such as police cars,
firetrucks, ambulances, and the rare times when a passenger
automobile has to break the usual laws due to an emgergency.
Finally, have you never heard of the acronym BSOD?
You don't even know what you don't know about how
computers work...or rather, don't work.
/BAH
jmfbahciv
<snip>
I suggest that you search for the times when a mere spider brought
down a comm system. And I won't mention those pesky little things
called squirrels.
/BAH
jmfbahciv
/BAH
jmfbahciv
> ji...@specsol.spam.sux.com wrote:
>> Yevgen Barsukov <evge...@gmail.com> wrote:
>
>>> Da! But that is not the main point. Main point is that they will
>>> have also much better ranging tools (lasers, radars, etc) which
>>> are always active, never look another way, never distracted.
>
> And hundreds of those radar devices in close proximity aren't going
> to cause a problem? They current ones are "confused" by the roadside
> furniture and have difficulty picking up things like pedestrians.
>
>>> This way robo-driver can not only "want" to follow the rules, it
>>> also "CAN" follow the rules much better.
>
> Fiction. Machines don't "want". They just do as previously
> instructed. No more. No less.
>
>> Absent a cable buried in the roadway, how would the car know where
>> it is in a lane?
>
>> GPS is not accurate enough for that.
>
> Lane-following using the lane markings.
>
> This isn't perfect; wet roads, snow, sand and other debris can
> confuse the systems currently deployed as driving aids by BMW,
> Citroën, etc.
>
> Lane markings can further be "confusing" due to road works with
> temporary markings e.g. diverting around the work site.
Not to mention the occasional moose and the many deer who like
to use the convenient pathways.
/BAH
Yevgen Barsukov
Good development team has people who understand all this stuff -
physicists (or chemist, or biologists depending what the application
area is) as well
as coders. Ever heard of system engineers, applications engineers
(usually Ph.D. in the field), firmware engineers? Obviously everyone
has
its own role, and the designing complex software like one
we are talking about involves tens to hundreds of people.
Does not mean there will be no mistakes. They will be corrected,
and each one needs to be corrected only once and for all million of
machines out there. At the other hand I observe tendency of
overestimating how good humans actually are.
If you look closely at any area of human activity, in average they
suck at it. And suck for good physical reasons, not only because of
faulty software (which is also widely present).
And you have to painstakingly try to fix this faulty software in
every single case individually, using very indirect and not always
successful methods (such as defensive driving school, fines, jail),
while with software you can just make a field update for all devices
in the world simultaneously in few minutes.
> Furthermore, there are vehicles
> which have a need to break traffic laws such as police cars,
> firetrucks, ambulances, and the rare times when a passenger
> automobile has to break the usual laws due to an >emgergency.
Automatic vehicles only need to follow rules themselves, they would
not expect other vehicles to follow the rules.
This is a good practice for human drivers as well.
So cops can go ahead and break as many rules as they want.
>
> Finally, have you never heard of the acronym BSOD?
>
> You don't even know what you don't know about how
> computers work...or rather, don't work.
Every real time firmware has watchdog reset implemented on hardware
interrupt level. I know because I design such firmware. I don't know
why M$ people did not think of that...
Besides, if you look at system design of most DARPA vehicles, they
are running multiple computers in parallel
on separate interrupts and similar but not identical inputs, which
makes it less likely that a space particle
or Turing's theorem will make all of them hang simultaneously. And
even if they did, it goes into safe mode and simple low level
controller takes over and safely parks the vehicle on the road-side.
And yes, one out of million, or out of 10 million will crash. But
now we have 1 out of 77 crash (that is btw the life-time odds of dying
from transport related accident). So one out of million would be a
tremendous improvement.
Bottom line - bugs will be there, but human hardware and software has
bugs too, only these are not fixable. We are not choosing perfection
vs. imperfection, we are just choosing better of the two
imperfections.
Regards,
Yevgen
>
> /BAH
Yevgen Barsukov
wrote:
> j...@specsol.spam.sux.com wrote:
> >Yevgen Barsukov <evgen...@gmail.com> wrote:
> >> Da! But that is not the main point. Main point is that they will
> >> have also much better ranging tools (lasers, radars, etc) which
> >> are always active, never look another way, never distracted.
>
> And hundreds of those radar devices in close proximity aren't going
> to cause a problem? They current ones are "confused" by the roadside
> furniture and have difficulty picking up things like pedestrians.
You are forgetting that humans can be confused too.
That is why we have continuous improvements of road markings, all this
reflectors, bright lines, road-side
barriers etc. Everything that benefits human driver will benefit
robotic driver. There is really no fundamental differences between the
two from the "camera analysis" stand-point, but robotic drivers can
have much more inputs.
You can put enough laser scanners to continuously create 3
dimensional topographic map of 100 m radius surrounding area. In this
topographic map you can find the road just because it will be more
flat and straight.
It can also continuously compare this map with pre-loaded map from its
GPS navigation, to make sure that everything coincides and makes
sense.
Human can not do anything of the kind and has to capacity to improve
(unless we implant some stuff into his brain)
>
> >> This way robo-driver can not only "want" to follow the rules, it
> >> also "CAN" follow the rules much better.
>
> Fiction. Machines don't "want". They just do as previously
> instructed. No more. No less.
From this logic, neither do people. They have pre-loaded
set of instructions on every level, from cell level (DNA)
to organ level (peripheral neural system), to brain level
(instincts) to higher cognition level (human culture program). If you
closely examine which of your knowledge you invented by yourself, you
will find that it is 0.0001% of what you know. Everything else is
ready-made hevristics (basically program that you execute) developed
by 100 000 years of human civilization. You just load the program "as
is" because people of authority (mother, older sister, teacher, book
authors) told you so. People don't even invent how to tie their
shoes. And there is nothing wrong
with that - because all these algorithms are proven correct by
millions of people and millions of different situations and therefore
are highly effective.
Yes, there is element of freedom (random decisions) if
your pre-loaded instructions are ambitious, or predicted outcome is
not satisfactory. This is true for both humans and machines - nobody
prevents you from putting random generator into software for the same
purpose, that would allow it to randomly try different choices and
evaluate the result.
The problem is, this "free search" method is highly ineffective for
either humans or machines. It wastes a lot of time, and it is not
guarantied to converge. It should be
used only if no algorithm that is proven effective in the past is
available.
>
> >Absent a cable buried in the roadway, how would the car know where
> >it is in a lane?
> >GPS is not accurate enough for that.
>
> Lane-following using the lane markings.
>
> This isn't perfect; wet roads, snow, sand and other debris can
> confuse the systems currently deployed as driving aids by BMW,
> Citroën, etc.
>
> Lane markings can further be "confusing" due to road works with
> temporary markings e.g. diverting around the work site.
Same as above - human drivers are routinely getting confused about
this, otherwise you would not see all
the cars the went off the curb, specially at night.
But laser ranging does not care if it is night or day,
neither does RM radar.
Regards,
Yevgen
tj Frazir
i dont even want to see a driverless car untill we all have the robots .
Your remember robots dont you.
well driverless cars wount wire any better and the LAST defence is a man
with the foot that stomps the brakes .
Yevgen Barsukov
We have robots. I bet you can spot 10 of them if you look around.
Lets see - washing machine, TiVo, dish washer, 3 PCs,
cell phone, iPod, robotic vacum, air-conditioner,
sprinkler system...
What you will not see is any "universal" robots. Universal
means "can do everything, but suck at everything".
We don't need such robots, we already have humans for that.
Regards,
Yevgen
Jeff▲Relf
The price as gone up 7.6 percent each year, on average,
and 10 percent per year the last few years ― much more, actually,
if you count my eviction ( for housing homeless kids ).
The growing horde of poor people are raising my rent,
even as the price of yachts goes down.
The Fed completely ingores us poor people, of course;
so they're litterally printing trillions of dollars
to raise the price of yachts back up to where they used to be.
Where the Fed sees deflation, the poor see only more inflation.
The Fed is all about the yacht salesmen ― you ― not the poorest, me.
tj Frazir
take the 10.000 acre forest and the 5 million inyour pocket..
ill hand you the buyer for 50,000 ton ( cut a fire trail 100 feet wide
10 mile long ) with contract cutters.
Use the buyers 10 million bucks and give 3.5 million for the north
maine forest.
take the money
Jeff▲Relf
“ lets change that ”, you tell me, but
I didn't say that the Fed was doing anything wrong,
I was just bracing myself for even greater costs
― rent+utilities ( the cheapest I can find ) are 95 % of my expenses.
I don't understand the “ 3.5 megaDollar, 10 kiloAcre ” Maine forest,
the 5 megaDollars in cash, the “ 10 megaDollar, 50 kiloTon ” buyer,
nor the “ contract cutters ” ― I haven't the first clue.
Who cluelessly invests in random deals with anonymous people ?
You sound like Bernie Madoff ( Made Off ) selling his ponzi scheme.
jmfbahciv
There will not be any good people doing that work because you have
eliminated all danger. Nobody of that ilk will be interested in
that kind of work because it will be boring.
>
> Does not mean there will be no mistakes. They will be corrected,
> and each one needs to be corrected only once and for all million of
> machines out there. At the other hand I observe tendency of
> overestimating how good humans actually are.
> If you look closely at any area of human activity, in average they
> suck at it. And suck for good physical reasons, not only because of
> faulty software (which is also widely present).
Right. People's efforts suck so much that you are posting based on
the work of ~100K people over the last 50 years. No computer can
innovate. What you want to do is destroy all opportunities to
innovate. This will stop trade. That will call a complete collapse
of the civilization you are familiar with. Perhaps this is a good
thing because it will eliminate all of these kinds of stupidos.
>
> And you have to painstakingly try to fix this faulty software in
> every single case individually, using very indirect and not always
> successful methods (such as defensive driving school, fines, jail),
> while with software you can just make a field update for all devices
> in the world simultaneously in few minutes.
Man! Are you way out in la-la land. Computer: "Excuse me while
I reboot. Pay no attention to that pile of metal, blood, flesh,
and fire one mile ahead....system restarting....system restarting...
system restarting...".
What happens when the power grid shuts down?
>
>> Furthermore, there are vehicles
>> which have a need to break traffic laws such as police cars,
>> firetrucks, ambulances, and the rare times when a passenger
>> automobile has to break the usual laws due to an >emgergency.
>
> Automatic vehicles only need to follow rules themselves, they would
> not expect other vehicles to follow the rules.
> This is a good practice for human drivers as well.
> So cops can go ahead and break as many rules as they want.
Nope. With your scheme, no humans will be allowed to drive.
>
>
>> Finally, have you never heard of the acronym BSOD?
>>
>> You don't even know what you don't know about how
>> computers work...or rather, don't work.
>
> Every real time firmware has watchdog reset implemented on hardware
> interrupt level. I know because I design such firmware.
I sure hope I don't ride transportation using your firmware designs.
You cannot issue a reset in the middle of a catastrophe if the
situation requires the hard/software to keep running. There are
other situations where the hard/software should stop running leaving
the controls and decision making to the human.
> I don't know
> why M$ people did not think of that...
> Besides, if you look at system design of most DARPA vehicles, they
> are running multiple computers in parallel
and what is the delay to resolve a conflict? How much time does
it take those parallel computers to decide that the conflict cannot
be resolved? When this happens, which of those parallel computers
becomes the primary CPU?
> on separate interrupts and similar but not identical inputs, which
> makes it less likely that a space particle
> or Turing's theorem will make all of them hang simultaneously.
You should really think about deadly embraces and CATCH-22s. At
least with a simultaneous hang, the human will know to begin
making the decisions...if he has any time left.
>And
> even if they did, it goes into safe mode and simple low level
> controller takes over and safely parks the vehicle on the road-side.
and if there is no side of the road? What if there is already a pile
of cars and semis there? What if the Grand Canyon is there?
>
> And yes, one out of million, or out of 10 million will crash.
Oh, bullshit.
>But
> now we have 1 out of 77 crash (that is btw the life-time odds of dying
> from transport related accident).
Where in the universe did you get those odds?
>So one out of million would be a
> tremendous improvement.
>
Assuming your numbers are correct, that one in a million crash will
not be a single car accident but a pile of cars, trucks, loads, and
fauna.
> Bottom line - bugs will be there, but human hardware and software has
> bugs too, only these are not fixable.
The coders are people. So you are stating that those who write the
bugs cannot fix them and will never learn from their
mistakes.
> We are not choosing perfection
> vs. imperfection, we are just choosing better of the two
> imperfections.
Oh, nuts. You are choosing dumbing down all people and
a government who will dictate who will be allowed to
go where and when.
/BAH
Yevgen Barsukov
Why would it be boring? Development of complex systems with large
scientific payload
is as geeky activity as you can imaging, just like making satellites
and military fighter jets.
I am lucky to be a part of such group (nothing to do with
transportation), and
it is cutting edge challenge which looks different every day. Being on
the front line is not relaxing
and not for everyone, but it is definitely not boring.
> > Does not mean there will be no mistakes. They will be corrected,
> > and each one needs to be corrected only once and for all million of
> > machines out there. At the other hand I observe tendency of
> > overestimating how good humans actually are.
> > If you look closely at any area of human activity, in average they
> > suck at it. And suck for good physical reasons, not only because of
> > faulty software (which is also widely present).
>
> Right. People's efforts suck so much that you are posting based on
> the work of ~100K people over the last 50 years. No computer can
> innovate. What you want to do is destroy all opportunities to
> innovate. This will stop trade. That will call a complete collapse
> of the civilization you are familiar with. Perhaps this is a good
> thing because it will eliminate all of these kinds of stupidos.
Let's step back here. I keep emphasizing that everything we do
is 99% executing in our brain programms optimized by our predecessors.
All inventions and discoveries are such programs loaded into our
brains. Of cause they are made
by people, little by little. But lets look more closely, what a
"person" that makes discoveries
is.
Humanity is a transmission line that takes all 100 000
years worth of optimized software for survival in this world, and
reloads it into the brain of a new-born monkey.
That makes the human-monkey look magnificently powerful (and it is)
but all this power
exists in virtual space, it is the power of loaded software, not of
the monkey itself.
And yes, this 100 000 years optimized software can make more
discoveries. It
is 100 000 years optimized discovery machine. But again it is the
software that
makes these discoveries, not the monkey. This has been proven by lots
of examples
of human babies lost in the wild - they don't develop into anything
better than an animal,
and does not make any useful discoveries.
So let's be clear - when we say "I can make discovery",
"I" in this sentence means not the monkey but the 100 000 old software
(keeping in mind that it
is a software that runs on monkey brain hardware)
At the other hand, if we focus only on the characteristics that
differentiate particular human,
a particular subset of whole humanity software + unusual deviations of
a particular brain,
this makes a very tiny contribution. Tiny does not mean unimportant
for you or me (as it is only
thing we have) but still tiny compared to humanity. So it makes much
more sense to associate
yourself with software (which is truly awesome, and eternal) than
hardware (which is
un-remarkable and short-living).
> > And you have to painstakingly try to fix this faulty software in
> > every single case individually, using very indirect and not always
> > successful methods (such as defensive driving school, fines, jail),
> > while with software you can just make a field update for all devices
> > in the world simultaneously in few minutes.
>
> Man! Are you way out in la-la land. Computer: "Excuse me while
> I reboot. Pay no attention to that pile of metal, blood, flesh,
> and fire one mile ahead....system restarting....system restarting...
> system restarting...".
>
> What happens when the power grid shuts down?
That all is already happening in real word. Of cause update does not
start during operation. Of cause there is back-up copy that is being
used until the whole overwrite is finished. So of there is power loss,
during programming, it will switch back to back-up copy, and re-try
later.
Again, it is not like humans can not be distracted. There is 4 times
higher odds to have an accident if there is 2 people in the car
compared
with 1 person in the car case.
> >> Furthermore, there are vehicles
> >> which have a need to break traffic laws such as police cars,
> >> firetrucks, ambulances, and the rare times when a passenger
> >> automobile has to break the usual laws due to an >emgergency.
>
> > Automatic vehicles only need to follow rules themselves, they would
> > not expect other vehicles to follow the rules.
> > This is a good practice for human drivers as well.
> > So cops can go ahead and break as many rules as they want.
>
> Nope. With your scheme, no humans will be allowed to drive.
For long time both human and automatic driven vehicles will coexist,
as we already have cars with active cruise control and plain old
clunkers with manual transmission on the road.
But even after all cars will become automatic, it is no problem -
police
car does not drive without rules, it just uses a different set of
rules. So it
police car will run a "police car" software, or even switch between
the modes
if necessary.
>
>
>
> >> Finally, have you never heard of the acronym BSOD?
>
> >> You don't even know what you don't know about how
> >> computers work...or rather, don't work.
>
> > Every real time firmware has watchdog reset implemented on hardware
> > interrupt level. I know because I design such firmware.
>
> I sure hope I don't ride transportation using your firmware designs.
> You cannot issue a reset in the middle of a catastrophe if the
> situation requires the hard/software to keep running. There are
> other situations where the hard/software should stop running leaving
> the controls and decision making to the human.
For complex systems you would have different time-outs for different
actions.
But any real-time system has to have timing diagram showing max times
for any action that software takes.
And "reset" does not necessary means power-down. It can just kill one
particular thread that run out of max. time.
>
> > I don't know
> > why M$ people did not think of that...
> > Besides, if you look at system design of most DARPA vehicles, they
> > are running multiple computers in parallel
>
> and what is the delay to resolve a conflict? How much time does
> it take those parallel computers to decide that the conflict cannot
> be resolved? When this happens, which of those parallel computers
> becomes the primary CPU?
Have odd number of computers so that conflict is resolved by a direct
vote.
This method is instantaneous.
> > on separate interrupts and similar but not identical inputs, which
> > makes it less likely that a space particle
> > or Turing's theorem will make all of them hang simultaneously.
>
> You should really think about deadly embraces and CATCH-22s. At
> least with a simultaneous hang, the human will know to begin
> making the decisions...if he has any time left.
Human is a real-time system and has its hardware and software
limitations too (just like you pointed
out about the "time left"). It can have conflicts that need to be
resolved (do I drive over the cat or hit
the side barrier, do I go over the double-line or do you go straight
and make a u-turn etc, etc)
Waking up for deep thoughts also requires time that in many cases is
not available in situation, so
resulting in a crash.
> >And
> > even if they did, it goes into safe mode and simple low level
> > controller takes over and safely parks the vehicle on the road-side.
>
> and if there is no side of the road? What if there is already a pile
> of cars and semis there? What if the Grand Canyon is there?
Same dilemmas exist for human drivers, and there are situations
without a right answer.
We are not trying to solve these, we are just trying to do best in
situations that do have
a solution.
>
>
>
> > And yes, one out of million, or out of 10 million will crash.
>
> Oh, bullshit.
>
> >But
> > now we have 1 out of 77 crash (that is btw the life-time odds of dying
> > from transport related accident).
>
> Where in the universe did you get those odds?
See here:
http://www.nsc.org/research/odds.aspx
This one has a nice graph comparing all kind of accidents. Prevalence
of traffic deaths is striking:
http://www.anesi.com/accdeath.htm
>
> >So one out of million would be a
> > tremendous improvement.
>
> Assuming your numbers are correct, that one in a million crash will
> not be a single car accident but a pile of cars, trucks, loads, and
> fauna.
>
> > Bottom line - bugs will be there, but human hardware and software has
> > bugs too, only these are not fixable.
>
> The coders are people. So you are stating that those who write the
> bugs cannot fix them and will never learn from their
> mistakes.
No, it is another way around -bug in the software for car control that
runs in human
brain are not fixable once and for all (they are fixable on individual
basis,
based on personal experience, fines payed, accidents experienced
etc).
I don't want to say that human software is not adaptable in general.
It is.
This adaptation is called "civilization". But it is a long process
that takes many generations,
and it works well in some areas and less well in others.
One typical difficulty is with real-time situations that require high
reaction rate and precision.
Human software and hardware is not very good for such tasks, it has
not have enough
evolutionary time to adapt to it.
It is just one case where "large universal" software does not work
well,
and a "small specialized firmware" is needed. In bio-systems all
critical functions are controlled by
"small specialized" sub-units, as it should. But evolution of such sub-
units take millions of years.
The problem is that such "driving" subunit did not yet have time to
evolve in homo sapience.
So we should try to accelerate the process and make an external one.
>
> > We are not choosing perfection
> > vs. imperfection, we are just choosing better of the two
> > imperfections.
>
> Oh, nuts. You are choosing dumbing down all people and
> a government who will dictate who will be allowed to
> go where and when.
First about "dumbing" - it is the opposite, I want "dumb" routine
task like driving to be eliminated, giving more time to do something
more creative (and saving lives of a lot of creative people out there
who
die because they are deep in thought about something interesting while
a track
turned up from the side lane). Are you not happy that you don't have
to hand-wash all
your cloth, and you can write to news-groups instead?
And - who said anything about government? So far all automatic
driving systems have been developed by multiple competing private
companies and universities. Active cruise control has been
independently
developed by Toyota, Volvo, BMW, Mercedes - all using different
methods
and trying to beat one another. I don't see why that would change.
Regards,
Yevgen
>
> /BAH
hanson
I just did the handouts and I will hide in guam
>
hanson wrote:
Merry Xmas, you Old Fart. *** Why the hiding? ***
Apparently the creditors are after you, aren't they?
If so, then simply send'em a nice Xmas card that
says:
::::::::::::: Cash is short and Times are hard :::::::::::::
::::::::::::: Here is your fucking Xmas card :::::::::::::
>
Take care, TJ, and keep them grand tales coning.
Thanks for the laughs.... ahahaha... ahahahanson
tj Frazir
Ill just drop a 50,000 MT buyer on ya.
Take his money and he will take the 50,000 ton .
Buy: Wood Chip mixture of Aspen and Birch (harwood) to make
pulp..........., Netherlands - Noord-Holland Importers / Exporters -
TradeBoss.com - B2B Marketplace, Import Export, Business Portal, FREE
Business Website
Address:http://www.tradeboss.com/default.cgi/action/viewtradeleads/tradeleadid/168398/subject/Wood_Chip_mixture_of_Aspen_and_Birch_harwood_to_make_pulp_/
$ 4 million for a forest from 10000 to 20,000 accres.
Tell ya what jeff ,,buy it in washington or origon maybe but maine for
shure.
2 ways to do it..
use the Import Eexport bank of the united states ,,web page .
Or use the buyers money .
You sit and fuck offall day anyway .
Why dont you check the forest from space and pics and buy the one you
like.
Just TAKE the money.
do it ,,,get it over with ..a forest and 5 million bucks to play with
will do ya good.
notice i buy forest farms islands castels and load ships as off shore
storage I buy no stock and didnt let 9 billion sit in the bank and get
lost.
im up to 9 fucking billion now.
What i cant get spent i hand to cherities .
ive got 9 but gave 5 away on the way
Jeff▲Relf
What arrows ? Google shows no arrows, see:
Google.COM/groups?selm=Jeff...@Seattle.2008_Dec.18|11.14pm|G
I see it like this:
JeffRelf.F-M.FM/MyPost.PNG
“ I'll just drop a 50 kiloTon buyer on ya. ”, you told me; here:
TradeBoss.COM/default.cgi/action/viewtradeleads/tradeleadid/168398/subject/Wood_Chip_mixture_of_Aspen_and_Birch_harwood_to_make_pulp_/
That's Greek to me, I don't understand the business.
Only a fool would engage in business he doesn't understand
under the advice of an anonymous/faceless person.
I don't need 5 megaDollars ― not that badly.
Eric Gisse
[snip relf droppings]
Here we are, with relf saying "only a fool would engage in business he
doesn't understand", with relf posting on sci.physics - a newsgroup
about a subject he does not understand.