Will AEVs kill PRT?

[Tim J.]

Worth watching. https://youtu.be/y916mxoio0E

He starts his analysis of five disruptive technologies coming together - batteries, electric vehicles, autonomous vehicles, on demand transportation, solar power.
He predicts that by 2030 95% of vehicle miles travelled will be in autonomous electric vehicles (AEVs) and people will stop buying private cars because transport-as-a-service will be 10x cheaper. Peak oil is predicted for 2022. Only two self driving technologies will control 90% of the market. Space allocated for parking cars will become available for other purposes.

Where will PRT fit in all this? Aside from some specialised use cases I can’t imagine anyone investing in an expensive guideway when an AEV will give door-to-door service.

All for simple economic reasons.

Regards, Tim

[Jerry Roane]

Tim

Thanks for the video.  What is missing is the  disruption of the disruption.  Energy overrides these operational differences.  Tracks are lower energy than the present asphalt riding cars.  PRT on guideway in certain locations will use less energy.  In defence of PRT.  Self driving of a dual mode EV can still do PRT operations and also unhook and take you home.  AEVs are here so discussing when they might happen is a done deal.  What I disagree with is losing private ownership of your horse.  (Personal cabin).   No one will desire sharing germs and COVID-19 with everyone.  

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/transport-innovators/0207E9D9-7483-493B-9FBD-7368506948D4%40gmail.com.

[Oded Roth]

A very interesting lecture, yet:

-        He properly criticizes Mackenzie's past prophecies about the cell phones, but trust his own prophecies.

-        The private cars have made very impressive advance since Henry Ford developed the production line, but the history of the private car commenced on the days of Nicola Cugnot and others in 1769 and a very long period past since some pioneers perceived  the potential of steam engine to tract carriages.

-        Further, the train industry overpassed the car industry apparently due to focusing on an integrated solution of the engine and its infrastructure.

-        Lessons from the history of PRT, (Bruno Latour or Catherin Burke's books), may shed some insights to the future of autonomous vehicles.

One can look at autonomous vehicles as a kind of personal rapid transit that uses cheaper infrastructure than the traditional unique solutions.

 

Oded Roth

Transportationet.com

‫בתאריך יום ו׳, 23 באוק׳ 2020 ב-1:05 מאת ‪Jerry Roane‬‏ <‪jerry...@gmail.com‬‏>:‬

To view this discussion on the web visit https://groups.google.com/d/msgid/transport-innovators/CAEsQt-dfDuvoZywW9x40mfPYmx-CVQnT%3DxFnR%2BE_jdRZzV%2B93Q%40mail.gmail.com.

[stephenw...@bellsouth.net]

Typical conference sales pitch.  Call it "impressionistic".  It paints a big picture by drawing all the details inaccurately.    Case in point - his Tesla AutoPilot numbers are total BS.

WRT your question:  I see a finite window for PRT to succeed.  At some point, the development of AVs will close that window.  As you said, if i can eliminate the guideway why wouldn't i.  I still believe that window is about a decade long.  The AV problem is a lot harder to solve than the hype machine would have us believe, and we've started to hear those admissions from the industry itself over that past year or so. Once it is solved, the power footprint for all that processing will still need to be reduced so that AVs are energy competitive.  It will happen - there is economic incentive.  But it will take some time.

The big advantage that AVs have is they use existing infrastructure.  PRT requires new infrastructure, which means projects must be executed through a public process.  In the US in particular, that regulatory process has been captured by entrenched industry, effectively preventing technological progress.   But that AV advantage turns into a disadvantage.  Performance is limited by the roadway system, not by the vehicles.  So IF we could have a fair competition between AVs (shared guideway) and PRT (exclusive guideway), i still think PRT wins.  But only 1 player is being allowed onto the field.

[Palle R Jensen]

I think the scale of the climate challenge will make AV vehicles uninteresting.

They do not solve the problems with energy consumption. An electric car is heavy and it has the same air resistance as a normal car, so if all 10 billion people on planet earth get access to drive AV, the climate will collapse.

Radical changes are needed.

 

Prof. Charles Lave from University of California Irvine stated: "RUF is our only hope of luring people out of cars".

I was interviewed by the German car magazine: Auto Bild (read by 7 million germans). I stated that AV will disappear when the hype period is over.

 

DualMode using a guideway for the long part of the trip is so powerful a concept that it will win - sometime :-)

 

 

--

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 view this discussion on the web visit https://groups.google.com/d/msgid/transport-innovators/2c48930c-b7b8-4b0c-a103-a5e9927fae02n%40googlegroups.com.

[Jerry Roane]

Pallet

Tesla has one of the best aerodynamic drag coefficients on the market.  They do have a big frontal area.  Minor point.  

Jerry Roane

To view this discussion on the web visit https://groups.google.com/d/msgid/transport-innovators/1UWINMtWRU.14K1FX9I9hH%40ejers.

[Tim]

AFAIK there are two divergent solutions to self driving.  

    1. HARDWARE/ENGINEERING: An environment mapping process that attempts to "identify" all objects in the path (or potential path) of a vehicle using expensive sensors like LIDAR   This is Waymo and others approach and is operating as driverless but only in one suburb in Arizona because it needs very precise maps.
    2. SOFTWARE/ITERATIVE LEARNING:  camera (cheap) sensors feeding data into a neural net.  This is Tesla and comma.ai.

Waymo etc. will fail because they cannot create a perfect model and the alternate technology is just as, if not more, functional.   From what I know of Latour's book it could be applied to Waymo whose ambitious declarations ten years ago have yet to be realised despite spending upwards of $10b.  (I know of Burke's book but not her thesis so I can't comment).   

I don't think it applies so easily to the second approach.  Telsa now has about a million vehicles on the road with hundreds of thousands using auto-pilot daily.   Comma.ai is selling their hardware for only USD 1,000 and has only about 5,000 users but is competitive with all the other platforms in terms of functionality.  Cars in these two platforms are being used in a diversity of real situations, not controlled environments.   In this model every time there is a disengagement by one driver the entire ecosystem has the opportunity to benefit in the next OTA update.   Just as the rules in a railway safety manual can be traced back to an accident, the "rules" in these self driving problems are based on the exceptions.  The more cars there are, the faster the learning.  

----

Self driving vehicles only have to be twice as good as than the average human alternative to achieve widespread adoption.   Telsa's figures already indicate they are preventing or minimising accidents.  (IIRC they are 4x safer).   It could be widespread adoption takes place in China first where car ownership is less affordable and learning to drive is a poor waste of time without owning a car.  

However the insurance industries underwriters are likely to be the ones who actually determine their take up since financial liability is the only important metric of risk in a market economy. 

I recommending watching Lex Fridman's interview with George Hotz, the comma.au founder, if you haven't see it.  He asserts that Telsa will introduce visual driver monitoring (which comma.ai already has) before they get to self driving. Here's one excerpt. https://www.youtube.com/watch?v=lLSNfUG5NUI

WHAT ABOUT EXCLUSIVE ROW AND PRT?  ISN'T THE BORING CO/TESLA PROPOSING TO BUILD AN ATN IN LAS VEGAS?
Musk has tacitly realised the need for an exclusive ROW that PRT/ATN asserts is necessary by starting the Boring Co and building their first installation as a three station-system in Las Vegas that uses only their own cars.  The application to build a larger system in that city has all the basic characteristics of a full PRT/ATN.  Logically driverless vehicles will come first in these systems because they can sanitise the guideway and control the environment.  It will be an irony if the first general transit PRT/ATN is built by an auto-maker.

Regards,
Tim