Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

[Tim J.]

Hello,

I came across the Izmir Institute of Technology PRT Demostration project (https://sites.google.com/site/ceng314prt/). When I search YouTube there are plenty of examples of impressive, simple bots (Pololu 3pi Line Maze Solving Robot, http://youtu.be/mJV-KDqHgDQ which sells for US$100 fully  assembled.)   There are very simple line follower bots.  But if you search for "PRT scale model" in YouTube the results are, politely, less than impressive.  (The film of the Aerospace Corp. model, from the 1970s(!!), are better than anything else.)

After this I wondered to myself why no one has created a decent PRT scale model with a dozen or so small bots running at high speed in a line drawn network using some of these off the shelf bots?  I would have thought some of robot aficionados would take it on as a challenge given they wouldn't have to start from scratch as there are so many algorithms and concepts in the public domain.

I found some work on other PRT models (http://www.autoroadvehicles.com/downloads/atra-2013-01-13-control-system-bob-johnson-3.pptx) but it was just one vehicle.  The discussion of scale model I have seen in this forum from last year seemed to focus on what scale it should be or on recreating the motion control system for one vehicle (https://groups.google.com/forum/#!topicsearchin/transport-innovators/Skytraxprt/transport-innovators/6DDSL1S6C0M).  

I would have thought the real challenge is the issue of continuous communication between the vehicles and the line side controllers and a central "dispatch" controller.  I realise that this is quite complicated to implement so maybe I am expecting too much.

Regards,

Tim

DISCLAIMER: I'm a curious (and sceptical) observer of PRT.

[Dennis Manning]

Tim J.

 

Did you see any information on the Skyweb Express ( www.skywebexpress.com ) 20 vehicle table top PRT scale model. They might be willing to talk about the challenges of building a PRT scale model.

 

Den

--
You received this message because you are subscribed to the Google Groups "transport-innovators" group.
To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.
To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Tim J.]

Hi Den,

Yes, I had seen their video but the vehicles just queue behind one another then advance to the head of the queue and then move to the next queue. The track they have allows for more complex operation than that.  I can only assume that given they have posted only that vision on-line then that was the best they could achieve.

I suppose what I was expecting is something more closely approximating what exists in all the fancy CGI simulations - merge, de-merge, off-line stations, vehicles bypassing stations.

On Friday, 2 August 2013 14:06:49 UTC+10, Dennis Manning wrote:

Tim J.

 

Did you see any information on the Skyweb Express ( www.skywebexpress.com ) 20 vehicle table top PRT scale model. They might be willing to talk about the challenges of building a PRT scale model.

 

Den

Sent: Thursday, August 01, 2013 8:47 PM

To: transport-...@googlegroups.com

Subject: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

Hello,

I came across the Izmir Institute of Technology PRT Demostration project (https://sites.google.com/site/ceng314prt/). When I search YouTube there are plenty of examples of impressive, simple bots (Pololu 3pi Line Maze Solving Robot, http://youtu.be/mJV-KDqHgDQ which sells for US$100 fully  assembled.)   There are very simple line follower bots.  But if you search for "PRT scale model" in YouTube the results are, politely, less than impressive.  (The film of the Aerospace Corp. model, from the 1970s(!!), are better than anything else.)

After this I wondered to myself why no one has created a decent PRT scale model with a dozen or so small bots running at high speed in a line drawn network using some of these off the shelf bots?  I would have thought some of robot aficionados would take it on as a challenge given they wouldn't have to start from scratch as there are so many algorithms and concepts in the public domain.

 

I found some work on other PRT models (http://www.autoroadvehicles.com/downloads/atra-2013-01-13-control-system-bob-johnson-3.pptx) but it was just one vehicle.  The discussion of scale model I have seen in this forum from last year seemed to focus on what scale it should be or on recreating the motion control system for one vehicle (https://groups.google.com/forum/#!topicsearchin/transport-innovators/Skytraxprt/transport-innovators/6DDSL1S6C0M).  

I would have thought the real challenge is the issue of continuous communication between the vehicles and the line side controllers and a central "dispatch" controller.  I realise that this is quite complicated to implement so maybe I am expecting too much.

Regards,

Tim

DISCLAIMER: I'm a curious (and sceptical) observer of PRT.

--
You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innovators+unsub...@googlegroups.com.

[Jack Slade]

You can look at my operating bench model at  SKYTRAXPRT  on youtube.  If you can ignore my crappy handiwork( I built it with wood)  and actually look at how it operates,  I think it shows how one very simple station would work.  Only with 5 cars,  because I did not have the space for a complete system.

 

Jack Slade

--
You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[Dennis Manning]

That doesn’t sound like PRT scale model I’ve seen. Try:

 

http://faculty.washington.edu/jbs/itrans/SWE%20marketing%20intro.pdf

 

Scroll down to  Alpha System (page 6)

 

Where is the on-line post you mention. I don’t believe I’ve ever seen it?

 

Dennis

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[Rick D]

Tim,

Might it not be easier and more informative to build accurate simulations of a large system?

Once the vehicle, track side, and central control systems are validated, a large scale prototype to validate key components and functions.

Isn't this what Vectus did?

---

From: Tim J. <tje...@gmail.com>
To: transport-...@googlegroups.com
Sent: Thursday, August 1, 2013 11:47:14 PM

Subject: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

[Kirston Henderson]

It is often less costly to just build the full-size system.

[eph]

This approach works well if funding is available.  Make CAD drawings, computer simulations and build the design.  If you are trying to persuade investors on a limited budget, I think a working model instead of a video montage (real or CG) gets more respect.  Urbanaut built a scale model and got funding to build in Korea.  The fate of the system is unfortunate, but does not necessarily reflect the process.  I think it's harder to fake a working scale model.


F.

[Kirston Henderson]

That is precisely the reason that MegaRail® designed, built, and
tested a full-size MicroWay™ mass transit passenger car working demo
model. With it, we were to fully test the critical vehicle steering
and switching with the real hardware, including the essential portion
of a short test guideway. A single 49-ft section and full sized
elevated guideway was built to verify guideway strength and deflection
and the emergency escape walkway feature.

Kirston Henderso
MegaRail® Transportation Systems, Inc.

[Tim J.]

Hi Dennis,  

It's on their website (on the link you sent me under "Technology")  see:  http://www.taxi2000.com/images/movies/AlphaVehicles.wmv

On Friday, 2 August 2013 14:52:25 UTC+10, Dennis Manning wrote:

That doesn’t sound like PRT scale model I’ve seen. Try:

 

http://faculty.washington.edu/jbs/itrans/SWE%20marketing%20intro.pdf

 

Scroll down to  Alpha System (page 6)

 

Where is the on-line post you mention. I don’t believe I’ve ever seen it?

 

Dennis

[Tim J.]

By simulation I take it you mean "computer simulation".    What I was expecting was a test network with a fair number of vehicles (10+) with all that MECHATRONIC complexities involved in integrating the electrical, electronic and mechanical systems.  I don't think a full scale prototype would be cheaper than small robots for this.

I'm not doubting the necessity for full scale prototypes but if there is only single or a couple of vehicles this doesn't really test resolve the functioning of the SYSTEM.  My original curiosity was stirred by the fact that as the full potential of PRT is supposed to emerge only when there is a number of vehicles on a non-simple network then why are their no decent scale models of such operation.   It seemed to me superficially that there is not a technical barrier to constructing one.

Perhaps the Skyweb Express model is more sophisticated than their own video shows (that's what the description in their documents say p.6 http://faculty.washington.edu/jbs/itrans/SWE%20marketing%20intro.pdf)  but I would be interested to see it.  As it stands I think their video is under whelming.  (There are worse on YouTube from JPods).

Regards,

Tim

[Rick D]

Tim,

Certainly a VERY underwhelming video at first look. I'd agree that it looks jerky and seems to be vehicles going from one stopped line to another.

The jerkiness would be why I would steer away from miniature models. A computer simulation setup with separate computers simulating different aspects of the system would fully validate the control systems on as big a system as the computer memory and performance can handle.

I've seen transit system control centers, automated and manual train operation, set up and fully operational while the track is still being laid. The control room equipment simply communicated with computers simulating the movement of trains and track and signals. This capability is almost vital, because even once the system is fully operational and all hardware is installed and trains running, the simulation system provides a means to train control centre personnel by simulating accidents, failures, and other various incidents.

For automated systems, one computer running vehicle control software and another running track/system interface validates vehicle  system control. And again, you can validate failures, incidents, and emergencies for the control software that would be difficult to duplicate in real life.

For the hardware validation, go as big as the budget can allow (Vectus, Ultra, or MegaRail as example). All you need is a divergence, a merge point, a high speed section, a few curves, a station. Something like this: http://www.aar.com/tracks.php (but in PRT scale).

With that covered, regulators can validate controls and hardware, and clients know that the systems works, is reliable (run a 10,000 mile simulation on the test track), and is fully scalable.

 

---

From: Tim J. <tje...@gmail.com>
To: transport-...@googlegroups.com

Sent: Friday, August 2, 2013 2:27:18 AM

Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

[Rick D]

Tim,

I failed to expand on the scale model and simulation a little in my last post and so will continue here.

"What I was expecting was a test network with a fair number of vehicles (10+) with all that MECHATRONIC complexities involved in integrating the electrical, electronic and mechanical systems."

The problem is that the scale model of the vehicle shows you virtually nothing about the reliability or operation of the final vehicle.

The skyweb video isn't so bad once you understand that the lines are actually stations. But getting the little vehicles to behave correctly and the "MECHATRONIC complexities involved in integrating the electrical, electronic and mechanical systems" is almost  as complex as the full scale version but is almost immaterial to the final full scale prototype.

For example, instead of spending all that time on little vehicles, there could be one computer simulating all of the cars, track, signalling, and other system components, and another connected by WiFi simulating a single vehicle.

As discrete system electronics are designed/adopted they can be integrated in their final form into the simulation. That is use the real communications hardware, propulsion controllers, power supplies, etcetera.

Meanwhile the full scale prototype validates the physical performance of the vehicle that corresponds to the outputs in the simulator.

---

From: Tim J. <tje...@gmail.com>
To: transport-...@googlegroups.com

Sent: Friday, August 2, 2013 2:50:31 AM
Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

[Dennis Manning]

Tim J:

 

You wrote: Yes, I had seen their video but the vehicles just queue behind one another then advance to the head of the queue and then move to the next queue. The track they have allows for more complex operation than that. I can only assume that given they have posted only that vision on-line then that was the best they could achieve.

 

That is what confused me. The video isn’t very good. The queuing that takes place is on the off-line tracks only. So if you look closely you will see that the model shows merge and demerge. The on-line track portions in the model are deceptively short.

 

I agree there is room for much better PRT scale models. I think it’s just the normal barriers – time, money, the right people to do the work,  etc.

 

Dennis

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[eph]

What would be needed is a "reality check" third party simulator that took a vehicle model with all it's characteristics and all the output data from the proprietary computer simulation and just verified that all characteristics, physics, etc.  all "worked".  Then the computer simulation would be on a similar footing with a scale model.

Building a full scale prototype is easier in some ways because COTS either exist or they don't so the design matches reality which is not the case with small scale models.  The big expense is the infrastructure needed to test the prototype at speed and in merges and diverges.  Road based vehicles have an advantage there since they can use existing infrastructure to a great extent.

I'm a bit on the fence on this one.  Small scale also has portability whereas full scale requires a major installation.


http://www.urbanaut.com/Basic%20Principal%20and%20Technology%204.htm




F.

[Tim Joy]

Do any of the simulations include a reality check of the kind you propose?  That sounds eminently logical to me.

I read in the Aerospace Corp. review of the San Jose proposal they were concerned none of the vendors mentioned potential problems with wireless communication.

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

--
Regards,
Tim
tje...@gmail.com
--------------------------

[Jack Slade]

Only a fool would plan PRT safety using wireless data in a situation where fiber-optics can easily be used.  Radio can be interferred with...by faulty local equipment,  by thunderstorms,  or by sabotage.   I don't have such a high opinion of Aerospace as I used to have on the recommendation of people who knew their reputation 40 years ago.  Too many people have retired or died,  and their replacements just use their heads to hang hats on. 

 

Jack Slade

From: Tim Joy <tje...@gmail.com>
To: transport-...@googlegroups.com
Sent: Friday, August 2, 2013 11:47:17 PM
Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

To unsubscribe from this group and all its topics, send an email to mailto:transport-innovators%2Bunsu...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

--
Regards,
Tim
tje...@gmail.com
--------------------------

--

You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[Tim J.]

I don't mind the jerkiness (although there are ways of simulating mass with the motor controllers such as better throttles on model railways ["model railroads" to you Americans :-)])  

I just don't understand the point of creating a fleet of vehicles, running all sorts of test and then showing us ONLY that.

My initial curiosity about this question was prompted by the following paragraph from the blog http://openprtspecs.blogspot.com.au/2013/02/railbotics.html

Ever see the video of that old tape by Aerospace Corp?  That is more than a good primer on PRT. That model is one very cool toy…There are still thousands of model railroad enthusiasts out there who get great satisfaction building similar stuff, not to mention robot buffs.  Could model “railbots” (not to be confused with the game) ever gain a following?  After all, a working model of the SMART PRT platform would certainly be equally fun to watch and vehicle avoidance and routing strategies would offer programmers something truly challenging to chew on.

I remain to be convinced about PRT in widespread network but I thought this was a fair point.  I find all the computer simulations like watching "Avatar"; anything is possible in CGI.

Tim

[Eric Johnson]

SkyWeb Express only has this video posted with a small clip of their models merging. Is there another video of their model setup?

Eric

[Dennis Manning]

What is your reservation about PRT in a widespread network? If one can have a large scale baggage handling network what problems does PRT have that make scalability a show stopper?

 

Dennis

--
You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[Tim J.]

I don't think large scale baggage handling is equivalent to a "widescale network" as it exists in a controlled, sympathetic environment and solve the  problem of the speed of disembarking by throwing its "passenger" off down a slide without stopping.  It's a limited analogue.

I should have added the extra condition of widescale URBAN networks.   to justify their capital cost would have to operate at, or near, their theoretical capacity.    If the origin/destination load is not balanced, it will need to overcome the speed of loading and unloading (larger stations?) and for a system of 100s of vehicles all will need to operate faultlessly.

Back on the issue of airport baggage handling I was surprised to read about the failed Denver Airport system that they were BOTH a.warned at the start that it was impossible to achieve (both in the time frame or at all at the time) AND b. that the tug & trolley system could NOT support the requirements.  The first proved to be true and the second to be false.  

I read some of the Aerospace Corp. review of San Jose and it seemed obvious that the headway of 6 seconds was excessively conservative but I assume this was the best the vendors who replied to the RFI reported to them.  Even at half that (still conservative) a network should be able to handle the projected 2030 demand.  In that kind of application I would imagine it might be superior.   I haven't analysed the alternatives but I think PRT would have to prove itself in just such an installation.

Perhaps the Amritsar project will put all of this to the test but I have read that to overcome capacity constraints they plan to have extra guideways.

I suppose part of my reason for posing the question on scale models in this forum in the first place there seemed to be an obvious gap between the (hyperbolic?) claims and the `facts on the ground'.

More broadly I always bear in mind that there are significant social and political problems in all forms of transportation for which there are no technological solutions.

I'm sorry if I have gone on so lone but I've tried to give a complete answer.   I look forward to any replies.

Regards,

Tim

You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.

[Jack Slade]

Tim,  I want to tackle your second paragraph only.

 

Not so.  When cities build a streetcar or subway line they never expect to get a cent of capital costs back.  In addition,  they know they will always have to subsidize it annually,  and over 20 years the subsidy will be greater than the capital costs.

 

If they spend an equal amount of money building PRT,  even light to moderate traffic will break even at the farebox.  When nobody moves,  the system costs nothing...no drivers,  no energy, and users should still be paying less than existing modes charge.  I have calculated $0.15 per mile will produce profits.

 

Jack Slade

 

From: Tim J. <tje...@gmail.com>
To: "transport-...@googlegroups.com" <transport-...@googlegroups.com>
Sent: Sunday, August 4, 2013 12:22:39 AM
Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

From: Tim J. <tje...@gmail.com>
To: transport-...@googlegroups.com
Sent: Friday, August 2, 2013 2:27:18 AM
Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

 

Hi Dennis,  

 

It's on their website (on the link you sent me under "Technology")  see:  http://www.skywebexpress.com/

--
You received this message because you are subscribed to the Google Groups "transport-innovators" group.
To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.
To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.
To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Tim J.]

Hi Jack,

Is that $ 0.15 per passenger mile or per vehicle mile?  

If its $0.15 per passenger mile what is the load factor?  

This would cover all operating costs including power, maintenance, cleaning etc.?

Most public transport doesn't pay for itself directly but there is a positive economic return.  In most cases the government subsidy is realised in the increased value of residential and especially commercial real estate along the route.  The alternative of buses or cars & car parks would be unsatisfactory.   The systems I was thinking about have to carry the crush-load of morning and evening peak,  The waste in the off-peak is only justifiable by their peak capacity.

In my thought experiment if these types of systems were replaced with widespread PRT I see the constraints on PRT at peak times would tend make it inoperable, particularly with the asymmetry of travel .  Is this fair?  Are there circumstances where light-rail is preferable?

If you're right we should see PRT systems spring up in the modernising cities of Asia, South America, Africa, the Middle East and even the conversion of existing public transport to PRT.

Do you think the Amritsar plan will be a test of the paradigm?  They are quoting figures of up to 100,000 passengers per day.

Just to tie this back to the topic of the thread: the capacity issue is why I was curious to see a model simulating a network running at its limit.

Regards,

Tim

[Eric Johnson]

Tim,

Let me take a crack at what you say and ask. Long posts are not a problem as we are here to discuss innovation transportation.

While there are differences between PRT and baggage systems, both are involved in getting something from multiple origins to multiple destinations in a complex layout. These baggage systems prove that we can write software that handles this complexity well and quickly and that we can build these complex networks. I see many basic PRT functions being verified by these baggage systems.

PRT networks provide better coveraged at lower cost vs any other transit system. If we assume inclusive system costs are $15 Million per mile, operating & maintenance costs of 4%, debt service costs are 5% of the capital costs, ticket price of $2.50 per rider, then we need 1538 riders per mile to break even. In the Phoenix area, that would mean a ~10% trip share which seems very doable considering 2/3 of the folks would be within a 5 minute walk of a station in the covered area. I've attached an Excel spreadsheet that you can play with and see what your numbers would do to ridership.

Capacity is an interesting discussion since your point applies to every transportation type, yet few people serious quibble when the new road expansion is swamped when it opens up. The same is possible for PRT and the nice thing is it's fairly easy for PRT to expand around those choke points or increase guideway capacity by reduced headways, platooning, increasing per pod ridership numbers, etc. If PRT capacity is maxed out, it should be pretty easy to get new financing to expand.

Denver was not an impossible problem. It could have been fixed with good leadership & time to debug the system. The system didn't work because of poor decisions made at multiple points in planning, design, and construction. Then since it wasn't working airport leadership took the easy way out by blaming the system instead of fixing it.

Aerospace went with 6 second headways even though we have plenty of systems that have proved 3 seconds or less are safe and reliable (Cabintaxi, Morgantown, Ultra, etc). Plus they never considered the idea that system operation after 30 years would confirm lower headways as being realistic and safe too.

Amritsar has steady heavy traffic all day long for some 20 hours a day. There system can handle 50,000 people per day per direction since they are looking to pack each pod with 6 people. If we assume they average 5 people per pod and 6 second headways that's 3000 people per hour per direction or some 60,000 people per day. Station capacity & design seems to be the main issue to handle that many people.

We base our claims on what PRT can do on simple math (headway * pax per pod, station throughput, speeds, etc) and that we've had several PRT test systems that have separately proven the concepts with different software & hardware. Guideway construction, vehicle design & construction, station construction, vehicle communications, etc are well known. We can use existing software or write new programs to meet our needs. The only thing needed to make a new PRT functional is a test track program to verify integration & operations.

These are just some of the reasons behind why I'm excited about what PRT can do.

Eric

On Saturday, August 3, 2013 9:22:39 PM UTC-7, Tim J. wrote:

I don't think large scale baggage handling is equivalent to a "widescale network" as it exists in a controlled, sympathetic environment and solve the  problem of the speed of disembarking by throwing its "passenger" off down a slide without stopping.  It's a limited analogue.

I should have added the extra condition of widescale URBAN networks.   to justify their capital cost would have to operate at, or near, their theoretical capacity.    If the origin/destination load is not balanced, it will need to overcome the speed of loading and unloading (larger stations?) and for a system of 100s of vehicles all will need to operate faultlessly.

Back on the issue of airport baggage handling I was surprised to read about the failed Denver Airport system that they were BOTH a.warned at the start that it was impossible to achieve (both in the time frame or at all at the time) AND b. that the tug & trolley system could NOT support the requirements.  The first proved to be true and the second to be false.  

I read some of the Aerospace Corp. review of San Jose and it seemed obvious that the headway of 6 seconds was excessively conservative but I assume this was the best the vendors who replied to the RFI reported to them.  Even at half that (still conservative) a network should be able to handle the projected 2030 demand.  In that kind of application I would imagine it might be superior.   I haven't analysed the alternatives but I think PRT would have to prove itself in just such an installation.

Perhaps the Amritsar project will put all of this to the test but I have read that to overcome capacity constraints they plan to have extra guideways.

I suppose part of my reason for posing the question on scale models in this forum in the first place there seemed to be an obvious gap between the (hyperbolic?) claims and the `facts on the ground'.

More broadly I always bear in mind that there are significant social and political problems in all forms of transportation for which there are no technological solutions.

I'm sorry if I have gone on so long but I've tried to give a complete answer.   I look forward to any replies.

Regards,

Tim

[Dennis Manning]

Eric and Jack:

 

Pretty good responses to Tim J’s misgivings about PRT being able to handle widespread urban deployments. Jay: I’d like to add just a short comment. You stated: I should have added the extra condition of widescale URBAN networks. to justify their capital cost would have to operate at, or near, their theoretical capacity.

 

I think you make a series of unwarranted assumptions. To begin with there is wide range of yet unproven capital costs for PRT. Second there is also a wide range of what constitutes theoretical capacity which includes vehicle speed, size, and headway. Then there is a large difference of opinion on what the ridership will be on any given system. However, if you make some basic and I think reasonable assumptions as to cost, capacity. and demand as Eric has done you will find PRT will do nicely at large urban scale. It’s easy to be be pessimistic and attribute high cost, low capacity numbers to PRT to argue it won’t work, I suggest you look over the information on these parameters. Perhaps it will alleviate some of your misgivings, and at this point I’d ask what you are basing your thinking on as to cost, capacity, and demand?

 

Dennis

 

From: Eric Johnson

Sent: Sunday, August 04, 2013 7:56 AM

To: transport-...@googlegroups.com

Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

 

--

[Jack Slade]

Calculations were for vehicle miles.  However,  when only the first line is built I can forsee the necessity to move 3 people per vehicle during rush hour.  We only get densities like this when we force people to travel from a couple miles each side of the line to use the system.  When the network gets finifhed walking distances shorten,  and line densities of 30,000 per mile will be unherd of,  just a fading bad memory.  I don't expect line density to ever exceed 5,000 except perhaps at special events such as ballgames,  rallies, and some organized celebrations after a full network is in operation.

[Eric Johnson]

Here is a video about baggage handling at Amsterdam's Schiphol Airport. It shows conveyor belt, a PRT style system, and storage operations in a 13 minute video. 

https://www.youtube.com/watch?v=Cqejei5WNcw

Eric

On Saturday, August 3, 2013 9:22:39 PM UTC-7, Tim J. wrote:

[Jack Slade]

Good video.  Did you notice the size of those baggage-transfer containers.  They are bigger than my PRT cars,  and move straight into the cargo hold...by conveyor.  Only last week somebody tried to laugh at my PRT conveyor system for moving people.  Ha Ha.

 

Jack Slade

From: Eric Johnson <itse...@gmail.com>
To: transport-...@googlegroups.com
Sent: Sunday, August 4, 2013 5:22:57 PM
Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

--

[Tim Joy]

Hi Eric,

 

Is it right to say “Amritsar has steady heavy traffic all day long for some 20 hours a day”?  Surely that would be a unique situation, unprecedented in fact.  (I don’t know much about the Golden Temple or Sikh rituals so I am willing to concede that it may be the case).

 

Also you say “Station capacity & design seems to be the main issue to handle that many people.“  Isn’t that the million dollar question for PRT?  The real constraint is in the stations.  (Not a unique problem to PRT).  I think I read somewhere that the Amritsar plan is to duplicate the guideways in some areas to add capacity.

 

If you say “The only thing needed to make a new PRT functional is a test track program to verify integration & operations“ does that mean CabinTaxi, ULTra and Vectus have NOT shown PRT is functional?  (Excuse my ignorance but I am guessing you have a different design).

 

You said:

Amritsar has steady heavy traffic all day long for some 20 hours a day. There system can handle 50,000 people per day per direction since they are looking to pack each pod with 6 people. If we assume they average 5 people per pod and 6 second headways that's 3000 people per hour per direction or some 60,000 people per day. Station capacity & design seems to be the main issue to handle that many people.

 

We base our claims on what PRT can do on simple math (headway * pax per pod, station throughput, speeds, etc) and that we've had several PRT test systems that have separately proven the concepts with different software & hardware. Guideway construction, vehicle design & construction, station construction, vehicle communications, etc are well known. We can use existing software or write new programs to meet our needs. The only thing needed to make a new PRT functional is a test track program to verify integration & operations.

 

Tim

tje...@gmail.com

 

 

 

From: transport-...@googlegroups.com [mailto:transport-...@googlegroups.com] On Behalf Of Eric Johnson
Sent: Monday, August 05, 2013 12:57 AM
To: transport-...@googlegroups.com
Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

 

Tim,

--

[Tim Joy]

Hi Eric,

 

I’ll have to study your spreadsheet to get a better feel for it.  It’s impressive with a lot of information.

 

After an initial view the first thing that comes to mind is I would have thought elevated stations would cost more than your calculations would indicate.  Have you compared these costs with any similar structures?

 

Tim

---------------------------

From: transport-...@googlegroups.com [mailto:transport-...@googlegroups.com] On Behalf Of Eric Johnson

Sent: Monday, August 05, 2013 12:57 AM
To: transport-...@googlegroups.com

Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

 

Tim,

--

[Eric Johnson]

Tim,

Station building costs are unverified and on the high side. I based them on Phoenix retail commercial construction costs and so I believe them to be on the high side for an enclosed, elevated structure. My elevated station concept is minimalist with concrete prefab floors & stairs, a carport type roof, an elevator, kiosk, and minimal railings. The ground level station would only have a concrete floor, a roof, and a kiosk.  Others believe stations should be enclosed with heating & cooling, seating, etc. I disagree with that since we believe folks will be waiting less then 1-2 minutes for their ride at worst.

I like to see numbers and I was getting tired of beating up my calculator up for different scenarios, so I built this spreadsheet to save my cheap $3 calculator... go figure...  ;-)
Eric

[Bruff]

Tim, you wrote that in your thought experiment you saw "the constraints on PRT at peak times would tend make it inoperable"

Constraints? What 'constraints'? 

DaveB

 

[Jeff Davis]

Tim,

I think that you are asking the right questions and hope you get meaningful answers.  As an FYI, please review a discussion of the Brick Wall Stop and sub-BWS headways that I wrote for the Innovative Transportation Technologies website.

 

http://faculty.washington.edu/jbs/itrans/BWS-Davis.htm

 

 

In regards to system capacity, please note that the peak capacity numbers being quoted are not sustainable since vehicles need to diverge out of the mainline and merge onto the mainline.  Vehicles exiting and entering the mainline disrupt mainline traffic (potentially slowing it down depending on the diverge and merge speeds versus mainline speeds) and creating gaps.

 

Also note that if the vehicles are self-powered (i.e. onboard energy storage), the dwell times cannot be less than the energy replenish time, otherwise the onboard energy storage will become depleted.

 

In regards to the complete dwell time considering passenger interaction, it would be helpful to see a complete timing analysis, such as door opening time, passenger deboarding time, passenger boarding time, and door close time.

 

Jeff

 

From: Tim Joy <tje...@gmail.com>
To: transport-...@googlegroups.com

You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[Tim Joy]

Hi Dave

 

I think the primary constraints for PRT are in the stations with

a.       the unloading/loading time of passengers and

b.       vehicles blocking the station path (either by backing into their berth, backing out of their berth or waiting for passengers to alight or board).

 

The other constraint is how to promote ride sharing a peak times.

 

In an earlier post in this thread there is some discussion of the Amritsar proposal and the problems of stations there. 

 

Has anyone analysed in detail the modelling of stations by Aerospace Corp in the report on the San Jose ATN?  From my cursory reading they indicate that station capacity does not increase significantly above 5 or 6 berths.

 

Tim J.

---------------------------

From: transport-...@googlegroups.com [mailto:transport-...@googlegroups.com] On Behalf Of Bruff
Sent: Tuesday, August 06, 2013 3:05 AM
To: transport-...@googlegroups.com
Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

 

Tim, you wrote that in your thought experiment you saw "the constraints on PRT at peak times would tend make it inoperable"

Constraints? What 'constraints'? 

DaveB

 

--

[Tim Joy]

Hi Eric

 

Have you put the specification of the San Jose airport ATN into your spreadsheet? 

How does the cost compare?  (I have seen a cost of US$700 million which is more than your estimate for a 40 mile system).  Would the “earthquake proof” civil engineering and a “sun shade” make such a large difference?

 

Tim

tje...@gmail.com

---------------------------

From: transport-...@googlegroups.com [mailto:transport-...@googlegroups.com] On Behalf Of Eric Johnson
Sent: Tuesday, August 06, 2013 2:03 AM
To: transport-...@googlegroups.com
Subject: Re: [t-i] Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

 

Tim,

--

[Dave Brough]

Hi back, Tim

a.       the unloading/loading time  The best real world example I can use is the lineup of cabs at an airport or a hotel - or here in Las Vegas, a casino, where the man toots his whistle, a cab pulls up, one guy hands the tooter a C-note while he and two or three others slid in, in like 10 seconds, and they're off. Toot-toot. Except with PRT, it would be four or six people and no toot or C-note. 

b.       vehicles blocking the station path (either by backing into their berth, backing out of their berth or waiting for passengers to alight or board).

Depends on the system. With those that have a single line and thus have to wait for the front pod to depart and where someone handicapped or drunk holds up the works, yes, a problem. Others (example MIST-ER), that have the pod backing out and then forward-ho, perhaps a couple of seconds - but still, a problem. Preferably, the track splits so that all pods just pull forward, rejoin the secondary line, and then return to the main line.

But comparing a pod stop to LRT or subway is overdoing it, don't you think? The LRT/subway line is straight-line transit, A to B, with station platforms sometimes a thousand feet long. The pod stop is within the footprint of a bus stop. That's for starters. Another is that the pod network is scattered. 

Even at that, a 50-pax transit bus running at 15 minute intervals is only moving 200 pax/hr. A pod system could handle that with half-minute departures. 

Subways can carry a lot of people, but the physical space (the tunnel) taken up by a subway could carry 4 pods each direction. The platform space could easily handle multiple pods. The problem there would be that subway stations are usually much further spaced apart.

The other constraint is how to promote ride sharing a peak times.

How about a simple Ap ("iPod" - or is that already taken?) 

But hold the train. PRT's not the answer at all. The answer is something that drives you from home to destination, not stop to stop, and that uses road as well as guideway. The name's on the tip of my tongue... 

Dave

[Eric Johnson]

Tim,

The Heathrow and Masdar systems seem to have passenger load times under 10 seconds in many cases. This doesn't include kiosk time where they select destination and in the future payment.

Here is a video someone did up showing how a very large crowd could be quickly handled by a station with 9 berths inline.  http://www.youtube.com/watch?v=-wZk0LbBPq8

Remember, people are used to doing their own tickets for LRT, airlines, etc so I don't see much of a learning curve there. Plus we expect a lot of the ridership will be repeat users and thus very experienced with dealing with the kiosk functions & loading process. Newbies or just plain slow folks probably will only account for less than 5% of our riders during rush hour. So, I don't see it as a major concern.

Needing to backup is one reason I don't like the sawtooth station design used by Heathrow. I would design things for a straight thru to help speed things up. Plus it simplifies station operations and construction too.

Eric

[Eric Johnson]

Tim,

Since the San Jose Airport study is for a small 5 mile system, we know it'll cost more per mile than you typically see tossed around here. I would change some things around to simplify the layout.  As ARUP added several things that I simply don't understand why. The PRT elevators at the terminals, the huge Terminal A station, three guideways wide, etc substantially raised their costs. 

So just changing length to 5 miles in the spreadsheet, it calculates it as $22 Million a mile. Since it's CA with higher business headaches and earthquakes I would bump it up to maybe $30 Million per mile/$150 Million total as a rough & conservative estimate. Heathrow was built for 43 Million Pounds/$65 Million and it's about 3 miles in length for comparison.

Eric

[Jack Slade]

Tim:  if you count the # of stations on a given line I think you will find that there is also a limit imposed by the capacity of the main line.   Eg:  6 miles of mainline would have at least 24 stations,  and if each station adds 5 passengers per minuite then that is 7200 people per hour.  The mainline would have to be operating at half-second headways to accomodate this.

 

It must be obvious that If this many people want to use that route another line should be built,  and such ridership will surely provide the profits to build it.  However,  there is no need to chasnge the stations,  as the station is not the problem. 

 

I have a different station arrangement than the sawtooth method,  and it will make stations work better.

 

Jack Slade

From: Tim Joy <tje...@gmail.com>
To: transport-...@googlegroups.com

You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[Kirston Henderson]

   It seems to me that much of the Heathrow ULTra guideway was built at grade around the perimeter of the parking lot rather than much more expensive elevated.

[Eric Johnson]

Kirston,

I just measured it and only about 600 feet of guideway out of 4300+ feet is at ground level, the rest is elevated until the parking lot stations. The ground level is required by aircraft safety reasons in case a plane landed short it's better to hit a ground level structure vs a heavy elevated concrete truss one.
Eric

[Tim J.]

Here are two (2) scale models I hadn't come across before.

1. YouTube video of Modutram scale model,  Uploaded on 26 Jan 2011,  1:06:
    http://www.youtube.com/watch?v=r7L7zuB-tMU

2. Image of Beamways model from December 2009

   http://smallcomp.blogspot.com.au/2009/12/beamways-podcar-model-to-be-shown-at.html

Regards,

Tim J.

On Friday, 2 August 2013 13:47:14 UTC+10, Tim J. wrote:

Hello,

I came across the Izmir Institute of Technology PRT Demostration project (https://sites.google.com/site/ceng314prt/). When I search YouTube there are plenty of examples of impressive, simple bots (Pololu 3pi Line Maze Solving Robot, http://youtu.be/mJV-KDqHgDQ which sells for US$100 fully  assembled.)   There are very simple line follower bots.  But if you search for "PRT scale model" in YouTube the results are, politely, less than impressive.  (The film of the Aerospace Corp. model, from the 1970s(!!), are better than anything else.)

After this I wondered to myself why no one has created a decent PRT scale model with a dozen or so small bots running at high speed in a line drawn network using some of these off the shelf bots?  I would have thought some of robot aficionados would take it on as a challenge given they wouldn't have to start from scratch as there are so many algorithms and concepts in the public domain.

I found some work on other PRT models (http://www.autoroadvehicles.com/downloads/atra-2013-01-13-control-system-bob-johnson-3.pptx) but it was just one vehicle.  The discussion of scale model I have seen in this forum from last year seemed to focus on what scale it should be or on recreating the motion control system for one vehicle (https://groups.google.com/forum/#!topicsearchin/transport-innovators/Skytraxprt/transport-innovators/6DDSL1S6C0M).  

I would have thought the real challenge is the issue of continuous communication between the vehicles and the line side controllers and a central "dispatch" controller.  I realise that this is quite complicated to implement so maybe I am expecting too much.

Regards,

Tim

DISCLAIMER: I'm a curious (and sceptical) observer of PRT.

[Jack Slade]

Tim  I am not sure I have all the answers you want,  but we have to consider what was available at the time most of us began to think about new systems.  Robots were certainly available then.

 

That $100 cost is certainly news to me.  Are you sure they have not left off at least 3 zeros?

It has always been my aim to try to keep car cost as small as possible,  in order to build more without breaking the bank.  What most people tend to forget that the sale price of all cars are about 7 times the manufactured cost.  I am quoting that figure from the average price markup in the USA

( it said 10 times more for Canada).  I do not know figures for other Countries.

 

Google may have great plans: only time will tell what percentage of the population will ever afford one.  The present wage trend here is downward,  and I am not interested in inventing PRT for the rich only.  I know that would be a good business plan,  and nobody has ever accused me of being a Liberal or a Socialist,  and it is hard to plan to break even when selling a service to poor people.

On the other hand,  my pension is not really lucrative,  and wouldn't I look stupid if I got to build something good,  and then couldn't afford to ride in it?  It is just as if somebody gave me an old used B727,  and I couldn't afford gas for it.

 

Jack Slade

From: Tim J. <tje...@gmail.com>
To: transport-...@googlegroups.com

--

[Tim J.]

Hi Jack,

The $100 is was talking about is for a Pololu 3pi robot as shown below.  You can see it at this site: http://www.pololu.com/catalog/product/975

The YouTube video I cited in my first post shows the very first post (http://youtu.be/mJV-KDqHgDQ) shows the sophistication (with the right programming of course) of these small devices.  

I wasn't suggesting that EVERYONE should try to create a complex scale mode (it is a difficult task, perhaps a waste of time); rather I was wondering why NO ONE had.  The film of the Aerospace Corp.  scale model from the 1970s still remains the best example I have seen.

Regards,

Tim J.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innovators+unsub...@googlegroups.com.

[Bruff]

Tim J.

Forget about animations or going to all the work and expense of scale models. This is the best thing to come along since sliced streetcars. As a PRT demonstrator, it might even shut up Avidor (Maybe me, too, Jack). 

All you'd need is a PRT-looking cover for the bot and black tape laid over a map of a particular city, Add in the stops, and there we go. Even better to have it real-time, in the sense that users could add or subtract the people flow according to time and event. 

Let's get to work!

Bruff

[Dave Brough]

Does anyone know if the Pololu setup (or any other)  can utilize multiple bots on the same grid? I couldn't find anything in the videos. I did ask on their info line, so maybe in a day or so will be able to report.

Bruff

Dave Brough

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.

[Eric Johnson]

Bruff,

While it's technically possible, it's adding a fair amount of interaction challenges that are probably too much for this $100 bot.

Eric

[Tim J.]

Hi 

The only "input" to that Pololu robot is the downward facing infra-red reader at the front.

However you can buy a Wixel Programmable USB Wireless Module for US$ 19.95 (ex delivery). [SEE BELOW]  You could add to that another $25 for a "collision detection sensor".  For $200 per robot I think you would have enough hardware. (plus a PC and a black texta of course. :-D)

IMO: The processors are quite powerful and with that transceiver there should more than sufficient bandwidth & processor speed for test bed network with 10 vehicles.  (This is just an intuitive guess but watching that YouTube video of how the bot "remembered" the maze and then accelerated and slowed along the long links tells me they are quite capable).

http://www.pololu.com/catalog/product/1337

Regards,

Tim J.  

[Tim J.]

For those still interested in this thread the following YouTube video shows communication between 3pi robots with the Wixel transceiver mentioned above given coordinated action

Collective control of m3pi robots. http://www.youtube.com/watch?feature=player_embedded&v=OKvZENrv6Po

As far as I can tell from the description at http://www.pololu.com/docs/0J32/4 there is no central "dispatcher".

If you want to see multiple robots running on a single track at high speed see: http://www.youtube.com/watch?feature=player_embedded&v=fl0CJhPiEfY 

These just run into one another eventually but you can imagine some integration of the above two way communication to create a PRT simulation.

Tim J.

[Burford Furman]

Innovators,

Just a quick update. I'm working with a handful of students this semester to design a very simple, low-cost 'PRT' test track using 1/32 scale slot car parts from the Scalextric4Schools program (http://www.scalextric.com/scalextric-4-schools/). The 'vehicles' will likely be battery-powered and guided within two guideway 'walls' formed by sheet metal strips that are mounted on several sheets of plywood. [Picture the Vectus vehicle, but instead of four thin wheels rolling on top of parallel rails, four wide slot-car wheels on plywood. Similar guidance within the side walls of the track]. The goals for the effort are:

1. Develop a very simple track, so the students can focus on control of the vehicles (speed, following, merging, exiting, etc.), integration of sensors and microcontrollers, and fleet management.

2. Have a relatively transportable demonstration of PRT concepts to take around to city councils, schools, and other venues to educate the public about PRT

3. Develop engineers who can get experience with controlling autonomous vehicles and build a supply for the other work we are doing on PRT at SJSU for Silicon Valley

I'm open for any input or thoughts on the effort. As we get further along, maybe I can post some pictures.

Thank you.

Buff Furman

[Jack Slade]

I will get back to you with a longer post later.  What you describe is more like UltRa.

May I forward your message to somebody I am occasionally in contact with there,  and maybe they will answer also?

 

Jack

--

You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[eph]

Interesting.

The 2 "walls" are the basis for ULTra (at Heathrow Airport).  Sensors on each side help guide the vehicle using curbs (kerbs) to bounce waves off (can't remember the sensor type off hand).

The crux of PRT is station to station transportation.  Go to the nearest station, punch in the destination station, get in and wait for the doors to open again at your destination.  Stations are "offline" meaning other "cars" can bypass while another is stopped.  On the track pictured, this could simply be a parking zone and a passing lane.  More intricate station configurations exist, some like at Heathrow require backing up the "car".

Tricky things are merging with other pods from a station/parking, laying out the best path (or a path) to destination, vehicle redistribution so there is no waiting at a station etc...


F.

On Saturday, September 14, 2013 12:10:17 AM UTC-4, Burford Furman wrote:

Innovators,

Just a quick update. I'm working with a handful of students this semester to design a very simple, low-cost 'PRT' test track using 1/32 scale slot car parts from the Scalextric4Schools program (http://www.scalextric.com/scalextric-4-schools/). The 'vehicles' will likely be battery-powered and guided within two guideway 'walls' formed by sheet metal strips that are mounted on several sheets of plywood. [Picture the Vectus vehicle, but instead of four thin wheels rolling on top of parallel rails, four wide slot-car wheels on plywood. Similar guidance within the side walls of the track]. The goals for the effort are:

1. Develop a very simple track, so the students can focus on control of the vehicles (speed, following, merging, exiting, etc.), integration of sensors and microcontrollers, and fleet management.

2. Have a relatively transportable demonstration of PRT concepts to take around to city councils, schools, and other venues to educate the public about PRT

3. Develop engineers who can get experience with controlling autonomous vehicles and build a supply for the other work we are doing on PRT at SJSU for Silicon Valley

I'm open for any input or thoughts on the effort. As we get further along, maybe I can post some pictures.

Thank you.

Buff Furman

.

[Burford Furman]

Jack,

No problem. The more heads on this the better.

No, I'm thinking more like Vectus, because it is not 'steered' like ULTra. Sensing where you are in the trough adds another layer of complexity to controlling the vehicles. The vehicles in our simple system will be single-degree of freedom - along the the track, to keep things simple. Switching must be done, and I suspect we'll come up with something like Vectus, with wheels that engage the side of the guideway and guide the vehicle either off to a station or stay on the main line.

Thank you.

Buff

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

--
************************************************************************************
Burford J. Furman
Professor
Department of Mechanical and Aerospace Engineering
San Jose State University
One Washington Square
San Jose, CA 95192-0087
Phone: (408)924-3817 FAX: (408)924-3995 Email: Burford...@sjsu.edu
Web page: http://www.engr.sjsu.edu/bjfurman/
************************************************************************************
"We must remember that intelligence is not enough. Intelligence plus
character--that is the goal of true education." - Dr. Martin Luther King Jr., 1947

"Instrumental or mechanical science
is the noblest and above all others, the most useful..." - Leonardo da Vinci

[Burford Furman]

Yes, these are the kinds of things I'd like the students to be able to devote more time to rather than also having to deal with the complexity of the mechanics that would be part of a more realistic vehicle-guideway interface.

--

You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.
To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

[Jerry Roane]

Burford

Do you have access to a makerbot?  If you do perhaps I could download your students with the parts necessary to do a full on realistic looking guideway network with no excuses door to door travel running on solar power.  (last mile solution)  I could be convinced to do all the CAD design work to scale it to 1/32 or some other exact scale you prefer (1/10th??) .  My 1/8th scale model I use for presentations is too big for an airplane carry on so I would recommend slightly smaller than Barbie and Ken sized at 1/8th.  The maker bot dimensions would need to be known so the parts are cut up into small enough chunks that can be glued of jointed together again.  

Don't underestimate your students.  They may surprise you.  

  

Jerry Roane 

--

You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[Burford Furman]

Jerry,

We DO have access to a 3D printer. I'll be happy to share with the students whatever you'd like to send.

Thank you!

Buff

[Kirston Henderson]

We built and tested a 1/20th working, dualmode scale model of our MegaWay™ vehicle very early in our development process to validate our guidance and switching method.  This working model was built from radio controlled car kits from a hobby shop plus a lot of custom designed and built parts.  It tested on line steering and switching plus off-guideway steering.  Fortunately, our engineer that designed and built the car was a pretty good machinist and had access to a small mill.  Now, a Makerbot could not make these parts at much lower cost.

I don't believe that your slot car guidance approach will be acceptable when it comes to switching.  I suggest that you use some sort of side rail sensing approach for your steering because the Vectus side roller approach to steering tends to provide rather rough steering, especially as speed increases.  I fear that if you have your students work with this type of steering, you will be teaching them a bad steering approach.

Kirston Henderson

[eph]

Oh, I think I understand (maybe).

So switch points can be a challenge.  I assume you will not be using slots, but if so, Jerry R. has a picture of a switch that could work.  Inter-vehicle spacing will be limited, but this isn't an issue for this sort of project.

A neat way to do a switch is to put metal plates on each side of the vehicles (or guiding feelers) and electromagnets on either side of the switch point.  Then switch the magnet(s) as needed to guide the vehicles one way or another.  Now your vehicles "only" need throttle control (wireless + batteries?) and switching is done from the guideway.  If computer control is the goal, vehicle position sensors in the track will be needed also or a cheap camera can be mounted overhead to sense vehicle positions (maybe reflective tape that gets obscured as a vehicle passes).  Lots of possibilities...  Lots of fun.


F.

On Saturday, September 14, 2013 1:25:45 AM UTC-4, Burford Furman wrote:

Yes, these are the kinds of things I'd like the students to be able to devote more time to rather than also having to deal with the complexity of the mechanics that would be part of a more realistic vehicle-guideway interface.

On Fri, Sep 13, 2013 at 9:34 PM, eph <rhaps...@yahoo.com> wrote:

Interesting.

The 2 "walls" are the basis for ULTra (at Heathrow Airport).  Sensors on each side help guide the vehicle using curbs (kerbs) to bounce waves off (can't remember the sensor type off hand).

The crux of PRT is station to station transportation.  Go to the nearest station, punch in the destination station, get in and wait for the doors to open again at your destination.  Stations are "offline" meaning other "cars" can bypass while another is stopped.  On the track pictured, this could simply be a parking zone and a passing lane.  More intricate station configurations exist, some like at Heathrow require backing up the "car".

Tricky things are merging with other pods from a station/parking, laying out the best path (or a path) to destination, vehicle redistribution so there is no waiting at a station etc...


F.

On Saturday, September 14, 2013 12:10:17 AM UTC-4, Burford Furman wrote:

Innovators,

Just a quick update. I'm working with a handful of students this semester to design a very simple, low-cost 'PRT' test track using 1/32 scale slot car parts from the Scalextric4Schools program (http://www.scalextric.com/scalextric-4-schools/). The 'vehicles' will likely be battery-powered and guided within two guideway 'walls' formed by sheet metal strips that are mounted on several sheets of plywood. [Picture the Vectus vehicle, but instead of four thin wheels rolling on top of parallel rails, four wide slot-car wheels on plywood. Similar guidance within the side walls of the track]. The goals for the effort are:

1. Develop a very simple track, so the students can focus on control of the vehicles (speed, following, merging, exiting, etc.), integration of sensors and microcontrollers, and fleet management.

2. Have a relatively transportable demonstration of PRT concepts to take around to city councils, schools, and other venues to educate the public about PRT

3. Develop engineers who can get experience with controlling autonomous vehicles and build a supply for the other work we are doing on PRT at SJSU for Silicon Valley

I'm open for any input or thoughts on the effort. As we get further along, maybe I can post some pictures.

Thank you.

Buff Furman

.

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.

To unsubscribe from this group and all its topics, send an email to transport-innovators+unsub...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

[Jerry Roane]

Buff

If you can tell me the height, width and depth of the printer I can do a printable version.  Depending on the printer these dimensions are quite different and there is some engineering on my part necessary to use standard off the shelf parts (bearings, shafts etc.) to make it work well at scale.  Scale models have to be beefier in certain areas and glue joints are just not as tough as the monolithic printed part.   The other bit of information is can you afford an extrusion die tooling for a plastic or aluminum guideway extrusion?  If you have enough in the budget to buy the tooling then your model can escape the plywood boundary and go yard scale.  Perhaps hamster sized for the customer base.  Done right this could go viral and the students could really get into it.    

Jerry Roane 

[Richard Gronning]

Buff,

In order to facilitate your student project, you should know that Taxi 2000 Corp, (T2C) has built such a project. They call it the, "Alpha Scale Model." They have 20 vehicles and 3 off-line stations. It is built on a slot car platform. They have a computer screen that shows the cars exactly as they move around the guideway. Perhaps T2C could aid your efforts. The engineer that laid it out no longer works for T2C, but he may be able to help as well.

See;

• Mike Lester <mle...@taxi2000.com>
• Don Mathews <math...@asme.org>

Best,
Dick

[WALTER BREWER]

There are at least videos of Jack Irving'd layout at The Aerospace Corp in the 1970's.

 

Try   thomas...@aero.org.

 

Walt Brewer

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

From: Richard Gronning

To: transport-...@googlegroups.com

[Burford Furman]

Jerry,

I'll find out about the dimensions and let you know.

Thank you.

Buff

To unsubscribe from this group and all its topics, send an email to transport-innovators+unsub...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

--
************************************************************************************
Burford J. Furman
Professor
Department of Mechanical and Aerospace Engineering
San Jose State University
One Washington Square
San Jose, CA 95192-0087
Phone: (408)924-3817 FAX: (408)924-3995 Email: Burford...@sjsu.edu
Web page: http://www.engr.sjsu.edu/bjfurman/
************************************************************************************
"We must remember that intelligence is not enough. Intelligence plus
character--that is the goal of true education." - Dr. Martin Luther King Jr., 1947

"Instrumental or mechanical science
is the noblest and above all others, the most useful..." - Leonardo da Vinci

--

You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innovators+unsub...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.

To unsubscribe from this group and all its topics, send an email to transport-innovators+unsub...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

--
************************************************************************************
Burford J. Furman
Professor
Department of Mechanical and Aerospace Engineering
San Jose State University
One Washington Square
San Jose, CA 95192-0087
Phone: (408)924-3817 FAX: (408)924-3995 Email: Burford...@sjsu.edu
Web page: http://www.engr.sjsu.edu/bjfurman/
************************************************************************************
"We must remember that intelligence is not enough. Intelligence plus
character--that is the goal of true education." - Dr. Martin Luther King Jr., 1947

"Instrumental or mechanical science
is the noblest and above all others, the most useful..." - Leonardo da Vinci

--
You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innovators+unsub...@googlegroups.com.

[Burford Furman]

Neat! As I recall the Aerospace scale model, they switched by turning on electromagnets at the switch point.

Thank you.

Buff

[Burford Furman]

Kirston,

Do you have anything documented about your MegaWay model that you could share with me?

I haven't seen the engineering details for the switching approaches used by Vectus, Taxi2000, and Cabintaxi, but from what I gather, they bear some similarity in that their bogies are guided within the guideway by side-facing wheels, and to execute the switch, some 'movable' wheel engages a surface on the guideway that urges the bogie to follow one way or the other. It would seem that all of these would have a similar problem of 'rather rough steering', is that right?

Thank you.

Buff 

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.

To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

[Burford Furman]

Dick,

Okay! Thanks for the contact info for Mike and Don.

Buff

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.
To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

[Burford Furman]

Walt,

Thanks. Jerry S. may have links to the videos on his site.

I have a call in to Tom.

Buff

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

[Kirston Henderson]

I don't have any documentation on the scale working model and I
believe that we discarded the test guideway and switch elements
somewhere along the way. We still have the working model sitting in
our Visitor Reception center. I will try to remember to take the
camera that we have and take some photos of it and when I do, I will
post them.

About every People Mover that I know of uses side guide-rollers,
except ULTra and 2Gethere. (That includes monorails and the Detroit,
Miami, and current DFW systems.) Because the side-mounted rollers
must have have some side freedom, they tend to shift the steering back
and forth as the vehicle moves. This shifting tends to create a cabin
side sway that can be annoying to passengers. It is similar to the
effect that occurs as the flanged steel wheels on railroad cars hit on
one rail and then the other, causing a constant rocking motion of the
passenger cabins.

As I understand it, ULTra uses a laser range finder to measure the
distance to one of the side rails and uses that for steering. I
believe that 2Gethere follows magnets embedded in the pavement.

Our systems use electomagnetic sensing of the distance to side
steering reference rails and uses this information for the steering
servos. The servo gain is set sufficiently high that the distance to
the side rails is essentially constant so that the cabins do not rock.

Kirston Henderson
MegaRail®

[Burford Furman]

Kirston,

Okay! Makes sense. Passive constraint (like flanged wheels on rail) trades simplicity for better ride quality of a more complex active guidance system that can be tuned to prevent oscillations.

Buff

[eph]

There is another related point to discuss which is the transition area needed by "passive constraint" switching.  This imposes certain guideway design limits not present in active switching.

On one hand, once the switch is made, safety is pretty much assured whereas with active constraints (like steering) can still fail.


F.

[Tim J.]

Hi Buff,

I think your scale model MUST include representation for the passengers in the systems.  Ultimately PRT is about the movement of "persons".  By including such an element in the design from the start it will help your students remain focused on the ultimate goal rather than just the motion of vehicles.  If your model is meant to be a convincing demonstration of PRT for lay people (and sceptics?) they need to keep this in mind.

Thus you would need:

 1.  a station "robot" that displays the number of passenger in the queue at each station and 

 2.  indicators on each vehicle would show its passenger load.  

(My preference would be, on the vehicles, would be to use an individual LED to represent each passenger since these will be easier to see on moving object.  An LCD counter could be used a the stations but an LED array could easily but used there.)

Your model could then visually demonstrate how PRT would respond to such challenges as asymmetry in demand, ride-sharing, average loads, loading times and the circulation of empty vehicles. These additions would entail trivial technical challenges compared to everything else but vital in making it a genuine scale model of a PRT system in operation.  

For position verification there are be cheap RFID readers and tags available and given the base of the vehicle (where the reading loop would naturally be) will be in close and fixed proximity to the track these should be able to read even at relatively high speed.  I don't have any direct experience with these (and the side-effect of interference from the motors etc) but I would guess you or your students will.

Regards,

Tim J.

[Jack Slade]

I think I have an answer that University students can  handle:

Go to any bus or other transit station in town,  and count how many passengers arrive,  and in what groups,  between busses or trains.

 

Then it will be easy to plan PRT for expected loads.  EG:  if carriers are 10 min apart,  and 10 separate passengers arrive during that period,  then with 1 vehicle every minuite there would never be a lineup.  It would also be an indication of what vehicle frequency would be needed in rush hours,  also,  to carry everybody with no delay.

 

In fact,  the PRT system would look like it is almost unused with that loading,  wouldn't it?

 

Jack Slade

[Tim Joy]

I think they would be better off using the demand matrix from the San Jose Airport as given in the Aerospace Corp. report.   Given the scale model guide-way will be limited in size to make it portable it would seem to me to be better to use passenger loads from an intensively used small system.  If it can handle high loads with asymmetry in demand then obviously it can handle anything less.

Alternately I would extend your example to say there are 5 stops each with 2 passenger/groups per minute.  That's now 10 vehicles per minute.  Assume they all want to get off at one or two destinations and there is a symmetrical back load so loading/unloading times come into play; I think that model would be interesting to watch.   Demand could then be incremented until performance starts to degenerate; the headway achievable in the scale models will probably be the main limiting factor.  I don't see the point of a scale model demonstrating it can solve a trivial problem.

Regardless of what demand and topology is modeled the main point I was recommending is that "passengers" must be included given they weren't mentioned in the original design specification. 

Tim J.

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

--
Regards,
Tim
tje...@gmail.com
--------------------------

[eph]

Tim, do you have a link to that demand matrix data?  I don't think I've come across it.

BTW, I like your idea of showing passenger counts in vehicles and queues in stations.  Some micro-controllers are also equipped with small screens.
I would like to see a station queue as a "list" of passenger groups.  So with LED, up to vehicle size LED on a row represent passenger groups and each row represents position in queue.  There could even be a queue for freight (or just a different colour) representing a cargo queue (maybe pallets).


F.

[Tim Joy]

Hi F. 

Below are links to two of the demand estimates from the Aerospace Corp report (p.24 and p.37 respectively) which says the source is Arup so I suppose they must come from there.

A link to the full report can be found at the bottom of http://www.sanjoseca.gov/index.aspx?NID=3706

City of San Jose home page "Home > Government > Departments & Offices > Departments & Offices P-Z > Transportation > Green Mobility >Automated Transit Network"

(I have highlighted a small error in the first table where, although it says "assumed symmetric", there is a missing term.  ;-))

Tim J.

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.
To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

--------------------------

[Richard Gronning]

The T2C Alpha model has lights on each vehicle that indicate passengers (or not).
The idea of passenger indication for stations is a good one.

Dick

[eph]

Thanks Tim,
Peak hourly demand is pretty low.  How in the world did Aerospace come to the conclusion no existing PRT system fit the bill?!  ANY PRT system can move those numbers of people.  325/hour is 11 s headway!


F.

On Sunday, September 15, 2013 9:20:28 AM UTC-4, Tim J. wrote:

Hi F. 

Below are links to two of the demand estimates from the Aerospace Corp report (p.24 and p.37 respectively) which says the source is Arup so I suppose they must come from there.

A link to the full report can be found at the bottom of http://www.sanjoseca.gov/index.aspx?NID=3706

City of San Jose home page "Home > Government > Departments & Offices > Departments & Offices P-Z > Transportation > Green Mobility >Automated Transit Network"

(I have highlighted a small error in the first table where, although it says "assumed symmetric", there is a missing term.  ;-))

Tim J.

-

[Rick D]

My question would be this: After all that work, what is the advantage over a good simulator?

Models are usually built to test and validate specific elements of a design or system. To ensure that a concept is viable, validate parameters, isolate a variable, test a component.

---

From: Tim J. <tje...@gmail.com>
To: transport-...@googlegroups.com
Sent: Sunday, September 15, 2013 12:11:27 AM
Subject: [t-i] Re: Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

Hi Buff,

I think your scale model MUST include representation for the passengers in the systems.  Ultimately PRT is about the movement of "persons".  By including such an element in the design from the start it will help your students remain focused on the ultimate goal rather than just the motion of vehicles.  If your model is meant to be a convincing demonstration of PRT for lay people (and sceptics?) they need to keep this in mind.

Thus you would need:

 1.  a station "robot" that displays the number of passenger in the queue at each station and 

 2.  indicators on each vehicle would show its passenger load.  

(My preference would be, on the vehicles, would be to use an individual LED to represent each passenger since these will be easier to see on moving object.  An LCD counter could be used a the stations but an LED array could easily but used there.)

Your model could then visually demonstrate how PRT would respond to such challenges as asymmetry in demand, ride-sharing, average loads, loading times and the circulation of empty vehicles. These additions would entail trivial technical challenges compared to everything else but vital in making it a genuine scale model of a PRT system in operation.  

For position verification there are be cheap RFID readers and tags available and given the base of the vehicle (where the reading loop would naturally be) will be in close and fixed proximity to the track these should be able to read even at relatively high speed.  I don't have any direct experience with these (and the side-effect of interference from the motors etc) but I would guess you or your students will.

Regards,

Tim J.

On Saturday, 14 September 2013 14:10:17 UTC+10, Burford Furman wrote:

Innovators,

Just a quick update. I'm working with a handful of students this semester to design a very simple, low-cost 'PRT' test track using 1/32 scale slot car parts from the Scalextric4Schools program (http://www.scalextric.com/ scalextric-4-schools/). The 'vehicles' will likely be battery-powered and guided within two guideway 'walls' formed by sheet metal strips that are mounted on several sheets of plywood. [Picture the Vectus vehicle, but instead of four thin wheels rolling on top of parallel rails, four wide slot-car wheels on plywood. Similar guidance within the side walls of the track]. The goals for the effort are:

1. Develop a very simple track, so the students can focus on control of the vehicles (speed, following, merging, exiting, etc.), integration of sensors and microcontrollers, and fleet management.

2. Have a relatively transportable demonstration of PRT concepts to take around to city councils, schools, and other venues to educate the public about PRT

3. Develop engineers who can get experience with controlling autonomous vehicles and build a supply for the other work we are doing on PRT at SJSU for Silicon Valley

I'm open for any input or thoughts on the effort. As we get further along, maybe I can post some pictures.

Thank you.

Buff Furman

--

You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[Jerry Roane]

Rick D.

This time we agree.  A good simulation would tell you and me everything we would like about the motion.  Problem is not everyone perceives the world the same.  To me when the core math is complete ---- job done.  Everything else is just a natural fallout of the core concept but --- not everyone sees the world that way and the model as expensive as it might be (example if a tiny scale model bearing costs $4.00 and you need X bearings it does get expensive) works for a large percentage of the population.  It you visit a future building site and you see one of those cardboard architectural models for the soon-to-be-erected building those too are expensive but they help sell the building and that is valuable.  The scale model, besides being a great Christmas tree toy, would help sell the concept to the general public.  If the architectural model did not pay for itself they wouldn't build them.  With 3D printing the labor and inaccuracy is greatly reduced.  Again the value to you and me of a printed part when the same part is there on the screen in the first place is less but to others it may be what gets them on board with the idea of small cars coming very often.  You can say the words but mostly people are not listening.  Build a scale system running on one solar panel and the energy concept may click on in their heads.  

Jerry Roane 

[Jeff Davis]

Jerry,

Perhaps you could define what is a 'good simulation' and how you propose to know or decide that the 'core math is complete ---- job done'?

 

Jeff

---

[Jack Slade]

Tim,  any report is usually just based on somebody's best guess.  I the report you mentioned,  so many facts were screwed up that I trust nothing in the document.  It is just hard to believe liars,  even when they tell the truth: problem is,  how do you tell?

 

I will fight like hell to keep this report from becoming a "standard" for any industry.

 

Jack Slade

To unsubscribe from this group and stop receiving emails from it, send an email to mailto:transport-innovators%2Bunsu...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.

To unsubscribe from this group and all its topics, send an email to mailto:transport-innovators%2Bunsu...@googlegroups.com.

To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

--
Regards,
Tim
tje...@gmail.com
--------------------------

[Jerry Roane]

Jeff

Here is one of many historical accounts of the core math.

E=M C squared

Manhattan Project

Emperor sun-god Hirohito realizes he is in fact not God.  

Jerry Roane 

[jbs]

Prof. Emeritus
U of Washington, Seattle
Innovative Transportation Technologies
URL: http://faculty.uw.edu/jbs/itrans

On Sun, 15 Sep 2013, Rick D wrote:

> My question would be this: After all that work, what is the advantage over a good simulator?

Many people (especially elected officials) are innumerate (can understand numbers). All the physical model can do is stimulate interest among those people and perhaps a desire for further investigation. Visual output from a simulation model would be more useful to some that can deal with graphs and charts. In my experience, I've seen many people who can't do that either, and that includes reading a map.

>
> Models are usually built to test and validate specific elements of a design or system. To ensure that a
> concept is viable, validate parameters, isolate a variable, test a component.

Yes, for the few that can deal with such results - absolutely necessary and valuable.
___________________________________________________________________________________________________________

>
> From: Tim J. <tje...@gmail.com>
> To: transport-...@googlegroups.com
> Sent: Sunday, September 15, 2013 12:11:27 AM
> Subject: [t-i] Re: Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of
> Technology PRT Demostration project)
>
> Hi Buff,

> I think your scale model MUST include representation for the passengers in the systems. ï¿œUltimately PRT is
> about the movement of "persons". ï¿œBy including such an element in the design from the start it will help
> your students remain focused on the ultimate goal rather than just the motion of vehicles. ï¿œIf your model

> is meant to be a convincing demonstration of PRT for lay people (and sceptics?) they need to keep this in
> mind.
>
> Thus you would need:

> ￜ1. ￜa station "robot" that displays the number of passenger in the queue at each station andￜ
> ￜ2. ￜindicators on each vehicle would show its passenger load. ￜ
> (My preference would be,ï¿œon the vehicles, would beï¿œto use an individual LED to represent each
> passenger since these will be easier to see on moving object. ï¿œAn LCD counter could be used a the

> stations but an LED array could easily but used there.)
>
>
> Your model could then visually demonstrate how PRT would respond to such challenges as asymmetry in demand,
> ride-sharing, average loads, loading times and the circulation of empty vehicles. These additions would

> entail trivial technical challenges compared to everything else but vital in making it aï¿œgenuineï¿œscale
> model of a PRT system in operation. ￜ

>
> For position verification there are be cheap RFID readers and tags available and given the base of the
> vehicle (where the reading loop would naturally be) will be in close and fixed proximity to the track these

> should be able to read even at relatively high speed. ï¿œI don't have any direct experience with these (and

[jbs]

Prof. Emeritus
U of Washington, Seattle
Innovative Transportation Technologies
URL: http://faculty.uw.edu/jbs/itrans

On Sun, 15 Sep 2013, eph wrote:

> Thanks Tim,

> Peak hourly demand is pretty low.ￜ How in the world did Aerospace come to the conclusion no existing PRT
> system fit the bill?!ￜ ANY PRT system can move those numbers of people.ￜ 325/hour is 11 s headway!

Because they (and ARUP) didn't include Cabintaxi as a candidate technology?

> F.
>
> On Sunday, September 15, 2013 9:20:28 AM UTC-4, Tim J. wrote:

> Hi F.ￜ

> Below are links to two of the demand estimates from the Aerospace Corp report (p.24 and p.37
> respectively) which says the source is Arup so I suppose they must come from there.
>

> A link to the full report can be found at the bottom ofï¿œhttp://www.sanjoseca.gov/index.aspx?NID=3706
> City of San Jose home page "Homeￜ>ￜGovernmentￜ>ￜDepartments & Officesￜ>ￜDepartments & Offices
> P-Zￜ>ￜTransportationￜ>ￜGreen Mobilityￜ>Automated Transit Network"
> Inline images 1

>
> Inline images 2
>
> (I have highlighted a small error in the first table where, although it says "assumed symmetric",

> there is a missing term. ￜ;-))
>
> Tim J.
> -

[eph]

They didn't include - cabintaxi, ultra, vectus and 2getthere?!

Of the not fully developed American ones - SkyWeb and Dr. Anderson's PRT?
SkyWeb claims headways of 1.1 s within APM rules.


F.

> transport-innovators+unsub...@googlegroups.com.

[Jack Slade]

You are probably right:  Therefore we should proceed with the course that proves the concept to the few,  and ignore the many? 

 

Then why is it that when we talk about Morgantown the whole scientific community stays silent,  and refuses to accept it as a fore-runner for anything.  By anything I mean capacity,  safety record,  mechanical reliability...the whole nine yards.  Instead of touting it as the World's first successful automated transit system,  it gets ignored.  You can drive within 2 miles of it on US79 without seeing one sign that it even exists.

 

When I stated that it's antiquated computer had done an excellent job for 40 years,  it was laughingly claimed,  on this list,  that I was proposing to use that computer to run my PRT system.

If that is not mis-representation,  or lying,  what is?

 

Jack Slade

From: jbs <j...@u.washington.edu>
To: "transport-...@googlegroups.com" <transport-...@googlegroups.com>
Sent: Sunday, September 15, 2013 1:55:49 PM
Subject: Re: [t-i] Re: Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

Prof. Emeritus
U of Washington, Seattle
Innovative Transportation Technologies
URL: http://faculty.uw.edu/jbs/itrans

On Sun, 15 Sep 2013, Rick D wrote:

> My question would be this: After all that work, what is the advantage over a good simulator?

Many people (especially elected officials) are innumerate (can understand numbers). All the physical model can do is stimulate interest among those people and perhaps a desire for further investigation. Visual output from a simulation model would be more useful to some that can deal with graphs and charts. In my experience, I've seen many people who can't do that either, and that includes reading a map.
>
> Models are usually built to test and validate specific elements of a design or system. To ensure that a
> concept is viable, validate parameters, isolate a variable, test a component.

Yes, for the few that can deal with such results - absolutely necessary and valuable.
  ___________________________________________________________________________________________________________
>
> From: Tim J. <tje...@gmail.com>
> To: transport-...@googlegroups.com
> Sent: Sunday, September 15, 2013 12:11:27 AM
> Subject: [t-i] Re: Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of
> Technology PRT Demostration project)
>
> Hi Buff,

> I think your scale model MUST include representation for the passengers in the systems.  Ultimately PRT is
> about the movement of "persons".  By including such an element in the design from the start it will help
> your students remain focused on the ultimate goal rather than just the motion of vehicles.  If your model

> is meant to be a convincing demonstration of PRT for lay people (and sceptics?) they need to keep this in
> mind.
>
> Thus you would need:

>       1.  a station "robot" that displays the number of passenger in the queue at each station and 
>  2.  indicators on each vehicle would show its passenger load.  

> (My preference would be, on the vehicles, would be to use an individual LED to represent each
> passenger since these will be easier to see on moving object.  An LCD counter could be used a the

> stations but an LED array could easily but used there.)
>
>
> Your model could then visually demonstrate how PRT would respond to such challenges as asymmetry in demand,
> ride-sharing, average loads, loading times and the circulation of empty vehicles. These additions would

> entail trivial technical challenges compared to everything else but vital in making it a genuine scale

> model of a PRT system in operation.  
>

> For position verification there are be cheap RFID readers and tags available and given the base of the
> vehicle (where the reading loop would naturally be) will be in close and fixed proximity to the track these

> should be able to read even at relatively high speed.  I don't have any direct experience with these (and

> the side-effect of interference from the motors etc) but I would guess you or your students will.
>
> Regards,
> Tim J.
>
> On Saturday, 14 September 2013 14:10:17 UTC+10, Burford Furman wrote:
>      Innovators,
> Just a quick update. I'm working with a handful of students this semester to design a very simple,
> low-cost 'PRT' test track using 1/32 scale slot car parts from the Scalextric4Schools program

> (http://www.scalextric.com/scalextric-4-schools/). The 'vehicles' will likely be battery-powered and

> guided within two guideway 'walls' formed by sheet metal strips that are mounted on several sheets of
> plywood. [Picture the Vectus vehicle, but instead of four thin wheels rolling on top of parallel
> rails, four wide slot-car wheels on plywood. Similar guidance within the side walls of the track].
> The goals for the effort are:
>
> 1. Develop a very simple track, so the students can focus on control of the vehicles (speed,
> following, merging, exiting, etc.), integration of sensors and microcontrollers, and fleet
> management.
> 2. Have a relatively transportable demonstration of PRT concepts to take around to city councils,
> schools, and other venues to educate the public about PRT
> 3. Develop engineers who can get experience with controlling autonomous vehicles and build a supply
> for the other work we are doing on PRT at SJSU for Silicon Valley
>
> I'm open for any input or thoughts on the effort. As we get further along, maybe I can post some
> pictures.
>
> Thank you.
>
> Buff Furman
>
> --
> You received this message because you are subscribed to the Google Groups "transport-innovators" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to

> transport-innovators+unsub...@googlegroups.com.

> To post to this group, send email to transport-...@googlegroups.com.
> Visit this group at http://groups.google.com/group/transport-innovators.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
> --
> You received this message because you are subscribed to the Google Groups "transport-innovators" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to

> transport-innovators+unsub...@googlegroups.com.

> To post to this group, send email to transport-...@googlegroups.com.
> Visit this group at http://groups.google.com/group/transport-innovators.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

--
You received this message because you are subscribed to the Google Groups "transport-innovators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to transport-innovators+unsub...@googlegroups.com.

[Jeff Davis]

Jerry,

Interesting choices for core math:

When A. Einstein wrote his paper with the now famous equation you cited, he was not sure if his mathematical analysis was correct.  He had sent it out for peer review.  He also discussed his analyses and logic with some very brilliant people who reviewed his work and provided insight and suggestions.  He also relied on others to provide the proof of his mathematical analyses.

Yes, the Manhattan Project employed a lot of very smart people who constantly peer reviewed each others work.

 

However, you did not explain how you propose to know or decide that the core math is complete ---- job done.

[Jerry Roane]

Jeff

When these mythical peers reviewed Einstein's work what letter or number did they add to the equation to make it complete?  Are you saying Einstein came up with 

E= -MC and his peer review process multiplied by minus one and another C after getting paid by the taxpayers as old sage consultants?  

Pretty good storytelling.  You should write our new history textbooks.  Revisionist 101.  "Young Einstein"  It could be a hoot like "Young Frankenstein"  

Best to stick to block control on a truck chassis.  

Jerry Roane

[Jeff Davis]

Jerry,

'Mythical peers reviewed Einstein's work'????  You're kidding right?

[jbs]

Prof. Emeritus
U of Washington, Seattle
Innovative Transportation Technologies
URL: http://faculty.uw.edu/jbs/itrans

On Sun, 15 Sep 2013, eph wrote:

> They didn't include - cabintaxi, ultra, vectus and 2getthere?!

Ultra, Vectus and 2getthere were on the list of proposals received by San Jose. Cabintaxi did not submit a proposal and was not on the list. Apparently, the consultants were told not to consider any company that was not on the list.

>
> Of the not fully developed American ones - SkyWeb and Dr. Anderson's PRT?
> SkyWeb claims headways of 1.1 s within APM rules.

As I recall, SWE was not on the list either. There were 8 vendors on the list but one had no PRT product to offer, just furniture. Another New Zealand company offered a suspended GRT product. Can't remember the other three.I posted the list a couple of weeks ago, I'll see if I can find it and post it again.
---------------------------

> On Sunday, September 15, 2013 1:59:27 PM UTC-4, jbs wrote:
>
>
> Prof. Emeritus
> U of Washington, Seattle
> Innovative Transportation Technologies
> URL: http://faculty.uw.edu/jbs/itrans
>
> On Sun, 15 Sep 2013, eph wrote:
>
> > Thanks Tim,
> > Peak hourly demand is pretty low.ￜ How in the world did Aerospace
> come to the conclusion no existing PRT
> > system fit the bill?!ￜ ANY PRT system can move those numbers of
> people.ￜ 325/hour is 11 s headway!
>
> Because they (and ARUP) didn't include Cabintaxi as a candidate
> technology?
>
> > F.
> >
> > On Sunday, September 15, 2013 9:20:28 AM UTC-4, Tim J. wrote:

> > ￜ ￜ ￜ Hi F.ￜ

> > Below are links to two of the demand estimates from the Aerospace Corp
> report (p.24 and p.37
> > respectively) which says the source is Arup so I suppose they must
> come from there.
> >
> > A link to the full report can be found at the bottom
> ofï¿œhttp://www.sanjoseca.gov/index.aspx?NID=3706
> > City of San Jose home page "Homeￜ>ￜGovernmentￜ>ￜDepartments &
> Officesￜ>ￜDepartments & Offices
> > P-Zￜ>ￜTransportationￜ>ￜGreen Mobilityￜ>Automated Transit
> Network"
> > Inline images 1
> >
> > Inline images 2
> >
> > (I have highlighted a small error in the first table where, although
> it says "assumed symmetric",
> > there is a missing term. ￜ;-))
> >
> > Tim J.
> > -
> >
> > --
> > You received this message because you are subscribed to the Google
> Groups "transport-innovators" group.
> > To unsubscribe from this group and stop receiving emails from it, send
> an email to

> > transport-innova...@googlegroups.com.

> > To post to this group, send email to transport-...@googlegroups.com.
> > Visit this group at
> http://groups.google.com/group/transport-innovators.
> > For more options, visit https://groups.google.com/groups/opt_out.
> >
> >
>

> --
> You received this message because you are subscribed to the Google Groups
> "transport-innovators" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to

> transport-innova...@googlegroups.com.

[eph]

I saw the list, but why the exclusion of other systems, then the "conclusion" that a "clean-slate" is required?!  Is it a fix or gross incompetence?

How does this happen or even get accepted as a document that has any merit?


F.

>       > transport-innovators+unsub...@googlegroups.com.

>       > To post to this group, send email to transport-...@googlegroups.com.
>       > Visit this group at
>       http://groups.google.com/group/transport-innovators.
>       > For more options, visit https://groups.google.com/groups/opt_out.
>       >
>       >
>
> --
> You received this message because you are subscribed to the Google Groups
> "transport-innovators" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to

> transport-innovators+unsub...@googlegroups.com.

[Tim Joy]

Hi F. 

Have you read the ATRA letter in response to the report?   (http://faculty.washington.edu/jbs/itrans/ATRA%20Response%20to%20SJ%20ATN%20Studies.pdf) 

That will answer your questions but I recommend you read Aerospace report itself and it will be obvious.

The must have got the 6 second headway from somewhere and one has to presume it came from the specifications received in the request-for-information (RFI) which in effect became the benchmark for state-of-the-art.  (If CabinTaxi didn't reply then it wasn't included)  It seemed obvious even to me as a sceptical bystander that this was very conservative but they are naturally risk-averse and they make the point themselves that even 1.6 second headway would still allow BWS safe-working.

Aerospace Corp. note pointedly that there is no PRT "industry" at present and it really akin to a "startup".  This is no doubt one of the reasons they give so many warnings about the risks entailed for the City of San Jose in an ATN "development".

Regards,

Tim J

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.

[eph]

Tim, yes, I've seen the documents.  Given the CLEAR misrepresentation of the state of the art, WHY is action being taken based on this garbage?

An even if the state of the art is essentially startup, with public service in 2 cases, how is the right course to start over?


F.

[Tim J.]

I'm the wrong person to answer but from what I've read they are pursuing the recommendations made by the City Of San Jose. That's why SJSU asked questions of this group. 

IMHO they want to develop a US industry rather than importing the off-the-shelf technology so a report that says "more work required" will give them the opportunity to do this. 

Are 6 second headways the best that ULTra and Masdar currently allow?  It seems very odd for ULTra given the projected capacity of the Amritsar ATN. 

Regards 

Tim J. 

[eph]

Tim, thanks.
ULTra: 4 s
http://www.prtconsulting.com/blog/index.php/tag/headway/

2getthere: 4 s
http://www.2getthere.eu/?page_id=58

Vectus: 3 s
http://www.vectusprt.com/uploads/518288B214A85.pdf

CabinTaxi: 2.5 s
http://www.advancedtransit.net/atrawiki/index.php?title=Cabintaxi

SkyWeb: 1.1 s (yet to be tested)
http://www.taxi2000.com/technology.html


F.

[Tim Joy]

Thanks F.

I hadn't see those figures before but they make Aerospace Corp.'s report look ridiculous; why didn't they even take headways at 4 seconds instead of 6 seconds?  They could have still made all their points about risks and complexities involved in developing an ATN without the biased analysis on capacity.  Bizarre!

I retract my previous speculation as to cause as it is now untenable.  [I suggest we go to Lewis Carroll to look for an answer. ;-)]

Regards,

Tim J.

[Tim J.]

Hi Rick,

I don't think a scale model is a substitute for simulations but I would put it this way: If 18 months ago a group of engineering students at SJSU had had a scale model of PRT running that consistently demonstrated 2 or 4 second headways without collisions would the Aerospace Corp. have dared to base its simulations on 6 second headways?

Furthermore the question would be: if this is what a group of under-graduates can do why can't we build it and why is Aerospace Corp going on about the development risks? It would form part of a proof of concept of PRT as a functioning SYSTEM, not simply relying on an extrapolation from the demonstrated properties of individual components.

The problem with simiulations is, as shown with CGI movies (e.g. "Avatar"), anything is possible. They function in idealised, abstract space A visceral demonstration would be qualitatively different since it entails interaction within physical space of a multitude of vehicles. Thus the `platform' the vehicles are based on is of secondary importance to the demostration of PRT behaviour. Ultimately the scale model is a substitute for the fact that there are no ATNs in existence (IMO Heathrow, Masdar, Suncheon & Morgantown and ALL the test tracks only demonstrate elements of the paradigm as classically described.) If Amritsar is built as described and works then there is no need for a scale model; the arguement would then be about the best design.

I might be in a minority but personally I would find it such a scale model pursuasive as on the other hand I find the disappointing performance of existing models to be dissuasive.

The thing that SHOULD dampen enthusiasm for the San Jose ATN is poor cost/benefit ratio given the low demand and high cost capital cost; but then again such a problem hasn't stopped many other public transport projects although they normally inflate their demand projections.

Regards,
Tim J.

[Burford Furman]

Tim,

Great ideas! Thanks for these suggestions.

Buff

On Sat, Sep 14, 2013 at 9:11 PM, Tim J. <tje...@gmail.com> wrote:

Hi Buff,

I think your scale model MUST include representation for the passengers in the systems.  Ultimately PRT is about the movement of "persons".  By including such an element in the design from the start it will help your students remain focused on the ultimate goal rather than just the motion of vehicles.  If your model is meant to be a convincing demonstration of PRT for lay people (and sceptics?) they need to keep this in mind.

Thus you would need:

 1.  a station "robot" that displays the number of passenger in the queue at each station and 

 2.  indicators on each vehicle would show its passenger load.  

(My preference would be, on the vehicles, would be to use an individual LED to represent each passenger since these will be easier to see on moving object.  An LCD counter could be used a the stations but an LED array could easily but used there.)

Your model could then visually demonstrate how PRT would respond to such challenges as asymmetry in demand, ride-sharing, average loads, loading times and the circulation of empty vehicles. These additions would entail trivial technical challenges compared to everything else but vital in making it a genuine scale model of a PRT system in operation.  

For position verification there are be cheap RFID readers and tags available and given the base of the vehicle (where the reading loop would naturally be) will be in close and fixed proximity to the track these should be able to read even at relatively high speed.  I don't have any direct experience with these (and the side-effect of interference from the motors etc) but I would guess you or your students will.

Regards,

Tim J.

On Saturday, 14 September 2013 14:10:17 UTC+10, Burford Furman wrote:

Innovators,

Just a quick update. I'm working with a handful of students this semester to design a very simple, low-cost 'PRT' test track using 1/32 scale slot car parts from the Scalextric4Schools program (http://www.scalextric.com/scalextric-4-schools/). The 'vehicles' will likely be battery-powered and guided within two guideway 'walls' formed by sheet metal strips that are mounted on several sheets of plywood. [Picture the Vectus vehicle, but instead of four thin wheels rolling on top of parallel rails, four wide slot-car wheels on plywood. Similar guidance within the side walls of the track]. The goals for the effort are:

1. Develop a very simple track, so the students can focus on control of the vehicles (speed, following, merging, exiting, etc.), integration of sensors and microcontrollers, and fleet management.

2. Have a relatively transportable demonstration of PRT concepts to take around to city councils, schools, and other venues to educate the public about PRT

3. Develop engineers who can get experience with controlling autonomous vehicles and build a supply for the other work we are doing on PRT at SJSU for Silicon Valley

I'm open for any input or thoughts on the effort. As we get further along, maybe I can post some pictures.

Thank you.

Buff Furman

--
You received this message because you are subscribed to a topic in the Google Groups "transport-innovators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/transport-innovators/ikdYsIDWCTc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to transport-innova...@googlegroups.com.
To post to this group, send email to transport-...@googlegroups.com.
Visit this group at http://groups.google.com/group/transport-innovators.
For more options, visit https://groups.google.com/groups/opt_out.

--
************************************************************************************
Burford J. Furman
Professor
Department of Mechanical and Aerospace Engineering
San Jose State University
One Washington Square
San Jose, CA 95192-0087
Phone: (408)924-3817 FAX: (408)924-3995 Email: Burford...@sjsu.edu
Web page: http://www.engr.sjsu.edu/bjfurman/
************************************************************************************
"We must remember that intelligence is not enough. Intelligence plus
character--that is the goal of true education." - Dr. Martin Luther King Jr., 1947

"Instrumental or mechanical science
is the noblest and above all others, the most useful..." - Leonardo da Vinci

[Jerry Roane]

Tim J.

Yes a scale model works with some but you still run into scale speed not being actual speed.  It has the same flaw as a simulation in that you can easily make a toy car go scale 300 mph but that doesn't mean anything at full scale.  The physical scale model is just one more representation of the simulation model.  What it does do is allow a larger slice of the population to visualize the interactions of the various parts.  You are also correct it does prevent the simulator from violating the physical laws like you can with CGI if you are not diligent.  My CGI representations are simulation driven but the viewer doesn't know that for sure where a plastic toy model at least you have to keep the movements within some very real physical constraints.  In the end you still have to build the rolling full sized unit.  Kirston has one.  I wish I had one.  The PRT vehicles in service of course are the ultimate but they still get detractors.  Seeing is not believing.  Believing is believing.    

Jerry Roane 

[Dennis Manning]

Tim:

 

I think you are right. The Silicon Valley wants to build an industry, so I guess this will be an ATN system designed by some kind of consortium. Good luck with a design by committee approach. They should just support the best American design, or maybe a couple of American designs, and get to building something ASAP. Endless studies by Aerospace is exactly the wrong approach. It will just widen the time lead of the Asians and Europeans.

 

Dennis

 

From: Tim J.

Sent: Sunday, September 15, 2013 10:47 PM

To: transport-...@googlegroups.com

Subject: Re: [t-i] Re: Why no decent scale models of PRT using off the shelf robots? (Izmir Institute of Technology PRT Demostration project)

 

You received this message because you are subscribed to the Google Groups "transport-innovators" group.
To unsubscribe from this group and stop receiving emails from it, send an email to transport-innova...@googlegroups.com.

[Kirston Henderson]

   Somehow, the sort of heavy manufacturing such as that needed to produce automated passenger vehicles and guideway does not appear to be a very good fit for a place such as City Of San Jose.  This sort of equipment is a long way from silicon chips, computer designs (built elsewhere), and software that seem to be the principal industry there.  Just what is the City Of San Jose going to get?

   In view of the tax structure and environmental and other restrictions in California, the type of industry needed appears to be fleeing the state.

Kirston