Friday, July 7, 2023

25th edition of the 11th year of SmartDrivingCars eLetter


TRB Automated Road Transportation Symposium
San Francisco, CA, July 9-13

Be sure to attend:
Tuesday, July 11, 7:00pm -> whenever

Alumni & Friends Banquet
Required: RSVP-Yes
Not required: Conference registration 
Wednesday, July 12, 1:30pm -> 5:00pm
Shark Tank Session
Required: Conference registration

Hmmmm…. I hope to see everyone in San Francisco next week, Alain


  SmartDrivingCars ZoomCast 324 / PodCast 324 Safe Street Rebel

F. Fishkin,  July 7, “”Safe Street Rebels” are disrupting robotaxis in San Francisco, Elon Musk says FSD is almost here…probably this year, TRB Symposium approaches, VW to test self driving in Austin and more. Tune in to episode 324 of Smart Driving Cars with Princeton’s Alain Kornhauser and co-host Fred Fishkin.

0:00 open
0:30 TRB Symposium on tap next week in San Francisco
6:45 TechCrunch headline: Robotaxi haters in SF are disabling Avs with traffic cones
15:00 California DMV AV collision reports
18:15 Musk again says Tesla is very close to achieving full self driving…probably this year
24:40 VW to begin testing self driving EVs in Austin
29:45 Zoox expanding from California to Nevada
39:50 Discussion of Techstination report: Should your car prevent accidents, period.


   Robotaxi haters in San Francisco are disabling the AVs with traffic cones

R. Bellan, July 6, “A decentralized group of safe streets activists in San Francisco realized they can disable Cruise and Waymo robotaxis by placing a traffic cone on a vehicle’s hood, and they’re encouraging others to do it, too. “

The “Week of Cone,” as the group is calling the now-viral prank on Twitter and TikTok, is a form of protest against the spread of robotaxi services in the city, and it appears to be gaining traction with residents who are sick of the vehicles malfunctioning and blocking traffic. The protest comes in the lead-up to a hearing that will likely see Waymo and Cruise expand their robotaxi services in San Francisco….”  Read  more  Hmmmm…   Can things get any uglier for Cruise and Waymo in SF?  Cruise and Waymo’s “SF Proof-of-market” “… These companies promise their cars will reduce traffic and collisions, … … has proven…   ”…but instead they block buses, emergency vehicles and everyday traffic,” reads one video posted on social media. “They even un-alived a person and a dog. And they’re partnering with the police to record everyone.”

This is nothing but ugly and not likely survivable in SF. The Cruise and Waymo technical pronouncements on congestion and safety signify that they were actually doing a “proof-of-technology” in SF.  Their focus on ride-hailers and enabling entitled Silicon Valleyers to party even harder are  messages that haven’t built an army of loyal customers are applying any counterweight to groups on the “safe street” high road.  A very poignant lesson is being taught here.  While Cruise and Waymo have proven that driverless cars can technically operate in the San Franciscos of this world, they have also proven that few, if any, will get to use them any time soon.  Time for them to pivot to “plan B” in search of a viable “proof-of-market”.  Alain

Should your car prevent accidents, period?

F. Fishkin, July 4, “ Would you want to own a car that would simply stop most accidents from happening?   What about having that kind of car for your children?   At Princeton University, the faculty chair of autonomous vehicle engineering, Alain Kornhauser, my co-host on the Smart Driving Cars podcast…says many vehicles today are equipped with enough technology or could be equipped with enough technology,  to simply not permit excessive speeding, tailgating and other forms of reckless driving and could prevent the vast majority of collisions…along with the associated deaths, injuries and costs.     The question to ponder is….is that something we want as a society?    The technology is ready and waiting.   The many who have suffered injuries or lost loved ones…would likely say yes.   What about you?   What about regulators and carmakers?   …” Read  more  Hmmmm…  Of course.  Fred and I have for years said there are 3 groupings of SmartDrivingCars:

*  SafeDrivingCars… exactly what Fred is talking about.  Their value proposition is they keep the driver from misbehaving if that misbehavior is likely to lead to a crash of any kind.

*  SelfDrivingCars… that perform the driving functionality when the driver remains engaged in overseeing the automated driving and remains completely capable of reengaging in the driving process within very short notice.  Their value proposition is the delivery of comfort and convenience to the driver.

*  DriverlessCars… that performs all of the driving functionality.  No assistance is required or desired to be done by any of the vehicle occupants.  These operate as well with or without any person in them.  Everyone inside is a passenger.  Their value proposition is purely an economic one in which no human labor expense is incurred in the provision of mobility.  This economic benefit can be profound in not only substantially reducing the cost of mobility but also enabling levels of service and vehicle utilization that are substantially better than can otherwise be achieved.  Alain


Staff, June 27, “As of June 27, 2023, the DMV has received 619 Autonomous Vehicle Collision Reports.

Collision reports prior to January 1, 2019 have been archived by DMV and are available upon request…”  Read  more  Hmmmm…   This link contains a sublink to the collision reports of each of the AV testers in California. The safety record to date is actually very impressive.  Just a few minor ones.  Nothing major. Alain

   Elon Musk says Tesla might achieve fully autonomous driving ‘later this year’

S. Schroeder, July 7, “It’s easy to forget amidst the Twitter/Threads drama, but Elon Musk is still CEO of that car company, Tesla.

And he’s still promising that fully autonomous, unsupervised self-driving is coming soon, even though he missed the mark on numerous occasions before.

Speaking at the 2023 World Artificial Intelligence Conference in Shanghai, China on Thursday, (via Electrek(opens in a new tab)) Musk said that he thinks Tesla is “very close to achieving full self-driving without human supervision” and will “achieve full self-driving, maybe what you would call (level) four or five, I think later this year.”… “.  Read  more  Hmmmm…   Emphasis is on the word “might” and who knows what is meant by “fully autonomous driving” .

 If “fully autonomous driving” =” Level 5”; “might’ = 10 to-the-minus-128 probability

If “fully autonomous driving” = a TrentonMOVES-style OOD that he focuses on starting July 15; “might’ = 0.25 probability


 Volkswagen will start testing self-driving cars in Austin as it moves on from Argo AI

J. Rosevear, July 6, “Volkswagen said Thursday that it will begin testing self-driving electric vehicles in Austin, Texas, later this month.

The German auto giant said it will deploy about 10 of its ID Buzz electric vans equipped with autonomous driving systems developed with Mobileye by the end of 2023. The first two of those vans are already in the U.S. and will begin testing before the end of July, it said….” Read  more  Hmmmm…   Wow!  Even though this is a “will begin testing…” announcement and not a “has begun…” this is the first good news in a very long time. A VW “will begin..” has to have  a better probability than an Elon “might achieve…”.  Alain

Zoox expands from California to Nevada, beginning in Las Vegas

S. Doll, June 27, “Less than five months after beginning completely driverless (and pedal-less) rides on public roads in California, Zoox has expanded operations to Nevada, beginning where else but fabulous Las Vegas. Nevada is quickly becoming a hub for robotaxis, but Zoox is captured a “first” in the Silver State thanks to its unique design.

Zoox is an autonomous vehicle developer celebrating closing in on nearly 10 years since its inception. In that time, former Intel chief strategist Aicha Evans has taken over as CEO, and the startup has secured a definitive merger with Amazon in which the online marketplace juggernaut wholly acquired it for an impressive $1.2 billion.

Contrary to nearly every other robotaxi company expanding on public roads, Zoox has developed a bespoke EV designed for the specific segment, free of a steering wheel and pedals….”  Read  more  Hmmmm…   More good news.  However, there is still an attendant on-board.  The real value of any of this only comes from the achievement of affordability through the elimination or having a permanent benefactor pick up the entire tab on the cost of an attendant.  Maybe Jeff Bezos ’86 is willing to create an annuity to pay for the attendant, now and forever. Although he may not be rich enough to meet the enormous demand by those that need a ride if high-quality affordable rides that these automated systems could deliver if only they could pass on the cost of the attendant to Jeff’s annuity.  Let’s see what happens in Las Vegas.  Lots of people need many rides every day.  Those who can afford Uber/Lyft/Taxi/StretchLimo are doing just fine thank you. Zoox has able competition.  Many folks in Las Vegas need a ride but can’t readily afford Uber/…  . Here Zoox could (without attendant or with Jeff’s largess) would have little competition and could thus deliver to these folks enormous value.  Let’s see which market Zoox goes after, and let’s monitor their success.  Alain

Tesla Owns Top Four Spots as Most American-Made Cars in Annual Report

R. Stumpf, June 21, “Each year, compiles the data of a study on auto manufacturing in the United States and publishes its in-depth results. More specifically, which cars are the most American-made, and how much they contribute to the overall U.S. economy?

The study compared 388 cars for the 2023 iteration of the American-Made Index (AMI). Despite the large sample size, only 100 made the cut as the most American vehicles, championed highest on the chart by Tesla and a significant number of foreign automakers.

Top 20 Most American Vehicles

  1. Tesla Model Y
  2. Tesla Model 3
  3. Tesla Model X
  4. Tesla Model S

Read  more  Hmmmm…  Wow!  Alain

   Feds propose automatic emergency braking systems for heavy-duty trucks

J. Gallagher, June 22, “Federal regulators are proposing a sweeping rule that would require all trucks over 10,000 pounds to be equipped with an automatic emergency braking (AEB) system and an electronic stability control (ESC) system that works in conjunction with AEBs.

The proposal, issued jointly on Thursday by the National Highway Traffic Safety Administration and the Federal Motor Carrier Safety Administration, would go into effect for most new Class 7 and 8 trucks (those with a weight rating of over 26,000 pounds) within three years of the final rule, with most new Class 3-6 trucks (weighing over 10,000 pounds) to meet the requirements within four years…. “  Read  more  Hmmmm…  It is about time!!!  Yea!!!  What about also mandating AEB that works for cars?  Alain

Self-driving firm Kodiak to equip 800 Loadsmith robotrucks

Reuters, June 24, Autonomous truck technology firm Kodiak Robotics will equip its self-driving software in 800 trucks for logistics platform Loadsmith, the companies said in a statement on Thursday.

Kodiak will start delivering the trucks equipped with Kodiak Driver in the second half of 2025, the companies said, adding that as part of Loadsmith’s U.S. freight network, the trucks will haul goods on the interstate highways while human drivers will complete local pickups and deliveries….” Read  more  Hmmmm…  OK, but this is a “gonna” that “starts” 2nd half 2025.  That’s 2 years from now.  I sure hope there are some up-front bookable payments on top of a complete stockout of inventory from today until that revenue starts flowing.  And that revenue is of order 10% of expected revenue flow in 2nd half 2025, else???? They desperately need an AEB mandate to survive.  Alain

Building for proximity: The role of activity centers in reducing total miles traveled

A. Tomer, June 29, “American households live amid a transportation conundrum. From a technological perspective, no developed country makes greater use of private vehicles and their incredible ability to cover long distances in relatively little time. The problem is that all those vehicles come at a real cost to society: growing environmental damage, unsafe roads, higher household transportation spending, and rising costs to maintain all the infrastructure. Even as electric vehicles promise to reduce the climate impacts of driving, this latest innovation still fails to address car dependency’s other persistent costs to society.

Building for proximity could offer a more holistic solution. Helping people live closer to the centers of economic activity—from downtown hubs to local Main Streets—should reduce the distances people need to travel for many of their essential trips. Shorter trip distances, in turn, make walking, bicycling, and transit more attractive and can improve quality of life. And as more people travel by foot instead of a private vehicle, officials can feel empowered to build complete streets that include lower speed limits, protected bike lanes, and other amenities.

In other words, greater proximity could lower environmental emissions, create safer streets, and unlock financial savings. …”  Read  more  Hmmmm…   Good report, but …

*   Today’s idyllic mode is a “15-minute” bike ride?  In today’s marketplace what % of 3 mile trips are being made by bicycle?  My guess is < 0.1%.

*   How is it that a household with one car saves X in car miles, and that same household with 2 cars saves 2X when in the first household that one car had to deliver all of the household’s mobility needs that were provided by 2 cars in the other household?  Didn’t that one car have to be used to give rides to the other household members and thus may have actually traveled more than the sum of the two cars in that 2-car household? The giving of rides to the miles traveled by by the non-driving household members is not explained.  By the way, how are the miles traveled by the driver counted in the provision of rides to the other household members?  Are those miles counted as traveled “purely” as a service provider to the mobility aspirations of the other household members?

*   I’m surprised that I could not find what % of US population lives in these “110 largest metro population.”  The closest I came was Table 1, which state that 3% of the population lives more than 11 miles from the nearest Activity Center (which I assume is in one of the “110”).  Is that really correct???  I’m confused.
*   Then in the appendix, the equation for household PMT has coefficients with a lead zero followed by  15 decimal digits (.3028684618342) to give estimates that “…  fall approximately 3% below the margin of error..” Why such precision in a denominator when the the numerator has, at best, single digit precision?  

Sorry, it is a good report, and land use is important; however, our mobility mess is a behavioral response by ~320 million individuals in what may well be the best democracy in the world.  It isn’t necessarily pretty, but are 15 minute bike rides really better?  That opportunity has been readily available for more than 100 years, and yet a precious few seem to have adopted it.

The aerial land-use image on the cover looks really nice, but its reality hasn’t led to its replication in any meaningful way.

Bridging Transportation Researchers (BTR) Conference

Paper Submission deadline: April 30

August. 9 & 10

On-line Conference


IATR 2023 36th Annual Conference