29th edition of the 9th year of SmartDrivingCars eLetter
CPUC, Nov 23, ’20, “This decision creates two new autonomous vehicle programs that authorize fare collection (deployment programs), one for drivered autonomous vehicles and the other for driverless autonomous vehicles. Among other requirements, applicants to the existing driverless pilot program and the new driverless deployment program must submit Passenger Safety Plans that outline their plans to protect passenger safety for driverless operations.
In addition, the decision establishes four goals that apply to both the existing pilot programs and the new deployment programs; 1.) Protect passenger safety; 2.) Expand the benefits of AV technologies to all of Californians, including people with disabilities; 3.) Improve transportation options for all, particularly for disadvantaged communities and low-income communities; and 4.) Reduce greenhouse gas emissions, criteria air pollutants, and toxic air contaminants, particularly in disadvantaged communities. The Commission will collect data to monitor permit holders’ progress toward each of the goals….” Read more Hmmmm… Sorry for not reporting this sooner, and thank you Doug Coventry for bringing it to my attention. It is must reading for any jurisdiction making regulations regarding the provision of autonomousTaxi mobility.
Its four goals are laudable, especially the 3rd, even if it may end up violating part of the 4th. Moreover, the clauses of the 3rd should be re-ordered to be: … Improve transportation options for disadvantaged communities, low income communities and those with disabilities, and, if possible, for all… This also reduces the goals to 3 important ones, … safety, the environment and improved mobility for those that have been left behind by the personal automobile
Of course, one wants to improve mobility for those that drive their own personal car; however, that is a entrenched well-served set of customers that are not readily going to flip from driving their car to something that isn’t really better and may largely be perceived as no cigar. Certainly, the public sector should in no way use public resources to give car drivers yet another good but inferior choice as was done with many public transportation investments that actually provide inferior mobility to those that were to be attract as customers. These systems are rebuffed by many that they were intended to be taken off the road for the trips they already make, let alone deliver quality-of-life benefits by providing mobility to new places that they couldn’t previously access.
A properly designed Operational Design Domain focused on from and where low income communities want to go is, to my mind, where the best opportunity exits for these safe, environmentally responsible systems . In such ODDs these driverless aTaxis can actually improve quality-of-life; and thus, deserve accommodation and promotion by public agencies such as CPUC. Alain
SmartDrivingCars Pod-Cast Episode 227, Zoom-Cast Episode 227 w/Ray Stern, news editor, Phoenix New Times
F. Fishkin, Aug 1, “https://youtu.be/Prun7fwOzYM In Arizona…the trial of a former Uber autonomous vehicle attendant om the death of Elaine Herzberg is still pending. Her attorneys say it is the company that should be responsible. From the Phoenix New Times, news editor Ray Stern joins Princeton’s Alain Kornhauser and co-host Fred Fishkin for a spirited discussion of the issues surrounding the case and more. Or you can listen to Episode 227 of Smart Driving Cars… https://soundcloud.com/smartdrivingcar/smart-driving-cars-episode-227-autonomy-responsibility-and-arizona” Alexa, play the Smart Driving Cars podcast!“. Ditto with Siri, and GooglePlay … Alain
The SmartDrivingCars eLetter, Pod-Casts, Zoom-Casts and Zoom-inars are made possible in part by support from the Smart Transportation and Technology ETF, symbol MOTO. For more information: www.motoetf.com. Most funding is supplied by Princeton University’s Department of Operations Research & Financial Engineering and Princeton Autonomous Vehicle Engineering (PAVE) research laboratory as part of its research dissemination initiative
Diana Furchtgott-Roth, Aug, 3. “Driving through upstate New York, every other vehicle appears to be a pickup truck such as a Ford F-150 or Chevy Silverado. But on August 11 the California Air Resources Board (CARB) will hold a public meeting to present draft regulations on requirements for Californians to purchase low- and zero-emission vehicles. CARB is charged with implementing California Governor Gavin Newsom’s 2020 Executive Order banning sales of new gasoline-powered passenger cars and trucks in the Golden State from 2035.
If California rules just applied to Californians, drivers in New York State and elsewhere would not have to worry. But rules that start in the Golden State drift east, like smoke from the wildfires. The Clean Air Act permits California to pass its own standards for new cars, subject to Environmental Protection Administration approval. Auto manufacturers like to produce for one market, so California has disproportionate clout….” Read more Hmmmm… Seems as if regulating the How (EV, ICE, Steam, Horse, …) instead of the What (pollution produced) is not good public policy. All one needs is to look at the latest California’s Lawrence Livermore (LL) Energy Flow Map to seriously doubt that banning ICEs in favor of EVs is going to deliver environmental benefits that are in line with the likely societal angst associated with the cure.
It remains questionable that any environmental benefits accrue from switching from an EV from an ICE. The LL Energy Map shows that 40% of California’s electricity is now generated from natural gas. I’m assuming that California operates its electrical systems so as to minimize environmental impact so that it is burning natural gas only because the other, less polluting, sources are maxed out. Thus, each new user of electricity in California, such as each conversion of an ICE to EV will be powered by natural gas. Moreover, two-thirds of the the generated electricity is lost (“rejected energy”) even before it gets to the car’s electric charger. So, one has to do some careful computations in the various scenarios to determine if powering personal cars with natural gas today in California is even infinitesimal better than with gasoline. (The answer is more obvious in states like Texas where almost 30% of electricity today is generated by coal. Not even close!)
For 2035 one needs to have a clear vision of how electric generation will evolve, how transmission losses can be reduced and how improvements in the ICE may emerge before one institutionalizes executive orders aimed at the How. Alain
N. Doodall, Aug ’21, “Most automobile manufacturers and several technology companies are testing automated vehicles on public roads. While automation of the driving task is expected to reduce crashes, there is no consensus regarding how safe an automated vehicle must be before it can be deployed. An automated vehicle should be at least as safe as the average driver, but national crash rates include drunk and distracted driving, meaning that an automated vehicle that crashes at the average rate is somewhere between drunk and sober. In this paper, automated vehicle safety benchmarks are explored from three perspectives. First, crash rates from naturalistic driving studies are used to determine the crash risk of the model (i.e., sober, rested, attentive, cautious) driver. Second, stated preference surveys in the literature are reviewed to estimate the public’s acceptable automated vehicle risk. Third, crash, injury, and fatality rates from other transportation modes are compared as baseline safety levels. A range of potential safety targets is presented as a guide for policymakers, regulators, and automated vehicle developers to assist in evaluating the safety of automated driving technologies for public use. …” Read more Hmmmm… This is a really good paper. It addresses the safety metric and in Table 2 summarizes the various safety measures across various modes, including elevators. As pointed out to me by Glenn Mercer, elevators don’t score very well as compared to many other modes; yet, few seem to hesitate using operatorless elevators.
My only issue is that the paper seems to focus on safety as being the only criteria in making decisions about AVs. Sure, safety is important, but there are other attributes of autonomousTaxis that are not irrelevant.
Buses may well be safe, but they offer lousy mobility and incur a high cost per passenger mile to deliver that lousy service.
Cars are pretty safe and offer fantastic mobility, but require substantial expertise that needs to be free for cars to be affordable.
Elevators aren’t all that safe but provide great up&down mobility without incurring a labor cost. Moreover, their capital costs are gladly covered by the locations that benefit (owe their very existence) from the mobility being delivered.
Anyway… very good paper. Alain
K. Horan, Aug 3, “A six-mile portion of I-24 will soon be the test site for a first-of-its-kind study on how autonomous cars impact traffic. Atop 110 foot tall poles, 300 ultra high definition cameras will view the section of road between Bell and Waldren roads. The goal is to collect data about how autonomous vehicles move in traffic and improve the flow of vehicles for everyone.
“Human drivers are actually less consistent than autonomous vehicles are today,” said Dan Work, engineer and researcher for Vanderbilt University. “So, we can actually pick up the nuances of the way that you or I drive that are distinct from the way automated vehicles drive…”Read more Hmmmm… A really good idea; however, unless some Teslas tend to drive this road segment, it will be a while before there may be enough “Self-driving” cars out there to measure anything but rare occurrences and a very long time before there are Driverless cars or trucks there.
What could be done with this system is to test the implications of cruise control, both the “unintelligent” (throttle only control) and the intelligent varieties. Vanderbilt could capture images of the existing flows and then ask TDoT to install a VMS to encourage drivers to use cruise control in their travels ahead. It would be very interesting to determine if the encouragement to use cruise control had any effect. This seems like an easy thing to do; however, I’m not aware that any state DoT has ever encouraged the use of cruise control in any manner. please let me know if I’m wrong on this. Plus, to make sense out of the results Vanderbilt will need to determine the extent of cruise control use in the before and the after recordings. They will also need to differentiate between non-intelligent and intelligent cruise control users. All no trivial details. Should be interesting. Alain
D. Meyers, Aug. 3, “The de Blasio Administration wants to force companies that road-test self-driving cars on the streets in spots around the five boroughs to apply for permits, according to recently proposed changes to city rules….” Read more Hmmmm… First thing to do is to ban them in Manhattan. Talk about the last Operational Design Domain that these vehicles should operate in. Also, the only reason anyone would do such a thing is for the buzz and not the substance. Since this is all about the buzz and not the substance, each of these companies need to be charged a very large fee for the use of the city’s streets for its promotional activities. The city’s streets are for mobility and not click bait. Alain
Elon Musk speaks out about Tesla’s $1.5 million payment to settle a battery-charging lawsuit, saying ‘if we are wrong, we are wrong. In this case, we were.’
K. Shalvey, July 31, “”If we are wrong, we are wrong,” he said on Twitter on Friday. “In this case, we were.”
Tesla agreed to pay $1.5 million to settle claims it had reduced the charging capacity on some vehicles in 2019, according to a settlement agreement filed in US District Court in San Francisco on Wednesday.
“Tesla policy is never to give in to false claims, even if we would lose, and never to fight true claims, even if we would win,” he said….” Read more Hmmmm… Very nice. Now he needs to change the name of his driver Comfort & Control systems and to insist that his owners remain alert, diligent and not mis-behave while these products are engaged. Alain
R. Stern, July 9, “No doubt, Rafaela Vasquez should have seen pedestrian Elaine Herzberg sooner on March 18, 2018, and taken action before the autonomous Uber vehicle she was riding in hit and killed her.
Widely seen interior video from a camera inside the Volvo SUV shows that Vasquez was not looking at the road in the seconds before the impact.
But there’s far more to the story than that, and Vasquez’s defense team says the grand jury didn’t get to hear information critical to the case before deciding to indict her last September on a charge of negligent homicide. Yavapai County Attorney Sheila Polk decided that Uber was not criminally liable in the crash in March 2019.
Her private lawyers, Albert Morrison and Marci Kratter, filed an extensive motion in Maricopa County Superior Court on Tuesday demanding that the case be remanded back to the grand jury for a new determination of probable cause….” Read more Hmmmm…In short my ethics say… Yes! See also Vasquez Remand Motion, July 9.
The algorithm “saw” Elaine 6 seconds before it hit her. The algorithm wasn’t written to side on caution … slowing down to take more time to resolve its confusion. The algorithm was written in such a way that it simply continued on “full steam ahead”. The algorithm had disabled the Automated Emergency Braking (AEB) system. The AEB was supposed to be explicitly deactivated only at speeds under 40 mph, yet the algorithm had the car traveling at 41 mph. Finally, the AEB itself may have been miscoded to explicitly disregard objects in the lane ahead for which the component of their speed in the direction of the lane centerline is sensed to be zero. Please don’t write code that does that!. Much of this miscoding by those that devise, chart and write these algorithms is out of a tendency to prefer comfort over safety/caution.
The act of driving down a road naturally involves the encounter with numerous objects for which “their speed in the direction of the lane centerline” is in fact zero. These are all of the stationary objects one encounters when traveling. Buildings along the side of the road, parked cars, telephone poles, picket fences, pedestrians waiting patiently for the light to change, etc. Unfortunately, the sensors that sense these objects, including LiDAR, are not perfect (nothing is), and will, while rarely, misplace these objects as being in the lane ahead. Moreover, there are stationary object that are indeed correctly sensed to be in the lane ahead, but these can readily be passed under… overhead signs, tree canopies and overpasses. Consequently, none of these stationary objects pose any danger. They can readily be passed under if they are really in the lane ahead and can be readily bypassed, if they are mis-located common stationary objects that line the road ahead… Unless it really is an object whose “speed in the direction of the lane centerline” is zero and it is really located in the lane ahead, as it was with Elaine Herzberg…. and with the rash of Tesla crashes with trucks sprawled across the lane ahead, firetrucks and police cruisers parked in the lane ahead, NJ barriers located in the center of an inappropriately striped exit lane, and trees in the lawn ahead.
Luckily, stationary objects in travel lanes are extremely rare, but, unfortunately, sensors and algorithms much more often mis-position objects in the lane ahead that are actually beside the lane, not in the lane. To avoid the “discomfort” of slowing down to be sure, these algorithms have been written to disregard, rather than be careful.
I my view, it is those that have written and implemented these algorithms that are the true folks that are “responsible” for this tragic crash. They didn’t have to write the algorithms that way. They could have written them to be better and more rarely mis-position stationary object. Moreover, they knew they had a problem here, because the code over-simplistically and irresponsibly dismisses its shortcoming. It is the way this code was written that caused this crash. The code required Rafaela to save it from this disaster. I doubt that Raphaela was informed about this fundamental shortcoming in the code.
Consequently, my ethics side that she is wrongfully charged. Whether or not the algorithm designers and coders need to be charged, is another question. They certainly should be aware that they are complicit here. So should the Society of Automotive Engineers who preaches “cause no harm’ and thus suggest that one never brake when one shouldn’t be braking. The person who is tailgating you may rear-end you. In a perfect world, then maybe. But, all of us, except for maybe SAEers, get confused, miss identify, mis locate and hopefully we all do hit the brakes at least a little to give us some time to get things straight. This philosophy should also apply to these automated gizmos. Alain
C’mon Man! (These folks didn’t get/read the memo)
Re-see: Pop Up Metro USA Intro 09 2020
K. Pyle, April 18, “It’s time to hit the start button,” is Fred Fishkin’s succinct way of summarizing the next steps in the Smart Driving Car journey. Fiskin, along with the LA Times’ Russ Mitchell co-produced the final session of the 2021 Smart Driving Car Summit, Making It Happen: Part 2. This 16th and final session in this multi-month online conference not only provided a summary of the thought-provoking speakers, but also provided food for thought on a way forward to bring mobility to “the Trentons of the World.”
Setting the stage for this final session, Michael Sena provided highlights of the Smart Driving Car journey that started in late December 2020. Safety, high-quality, and affordable mobility, particularly for those who do not have many options, was a common theme to the 2021 Smart Driving Car Summit. As Princeton Professor Kornhauser, the conference organizer put it,…..” Read more Hmmmm…. We had another excellent Session. Thank you for the summary, Ken! Alain
Ken Pyle‘s Session Summaries of 4th Princeton SmartDrivingCar Summit:
15th Session Making it Happen – Part One: Elected Officials’ Role in Creating a Welcoming Environment in the Trentons of this World
Kornhauser & He, April 2021 “Making it Happen: A Proposal for Providing Affordable, High-quality, On-demand Mobility for All in the “Trentons” of this World”
Orf467F20_FinalReport “Analyzing Ride-Share Potential and Empty Repositioning Requirements of a Nationwide aTaxi System“
Kornhauser & He, March 2021 “AV 101 + Trenton Affordable HQ Mobility Initiative“
Calendar of Upcoming Events
5th Annual Princeton SmartDrivingCar Summit
Live in Person
Tentaively: November 2 (evening) -> 4, 2021
R. Shields, 22 – 25 March, “Recordings from the conference:
Session 1 plus opening: (Regulatory): https://youtu.be/UcDC8gXiUFk
Session 2: (Cybersecurity): https://youtu.be/ppp2hxlvebY
Session 3: (Automated Driving Systems): https://youtu.be/uL2dRHuX2Cc
Session 4: (Communications for ADS) : https://www.youtube.com/watch?v=IFQcL6yfBso
Read more Hmmmm… Russ, thank you for sharing! Alain