26th edition of the 9th year of SmartDrivingCars eLetter
Neal Boudette, July 5, ” Benjamin Maldonado and his teenage son were driving back from a soccer tournament on a California freeway in August 2019 when a truck in front of them slowed. Mr. Maldonado flicked his turn signal and moved right. Within seconds, his Ford Explorer pickup was hit by a Tesla Model 3 that was traveling about 60 miles per hour on Autopilot.
A six-second video captured by the Tesla and data it recorded show that neither Autopilot — Tesla’s much-vaunted system that can steer, brake and accelerate a car on its own — nor the driver slowed the vehicle until a fraction of a second before the crash…” Read more Hmmmm… A few comments here:
1. Because of the suit here, hopefully more of the data associated with this crash will be made public. Future crashes such as these seem to be covered by the recent NHTSA standing General Order requiring the data to be released without need of lawyers, assuming Tesla cooperates.
2. Neal (slightly) overstate his plot which clearly shows the Tesla began to decelerate slightly more than a full second before impact. He also doesn’t mention what the video clearly shows that the Tesla was “cut-off” by the pickup truck. More over the pickup applied its brakes as it was making the lane change (brake light came on). This brake application may well have been the critical element that made the crash unavoidable. AutoPilot was likely tacking the pickup from at least the 6 seconds before collision point. Tesla must have data on the relative longitudinal speed between the pickup and the Tesla.and it must also have an expected time-to-collision which is a critical measure as to when to kick in the Automated Emergency Braking System. Once again, my main concern here is not (yet) about the performance of AutoPilot, but the performance of Tesla’s Automated Emergency Braking System (AEBS). This is a rear-end crash. It is the responsibility of the AEBS to avert these crashes. Seems as if the AEBS did NOT properly anticipate the pick-up’s maneuver nor properly monitor time-to-collision. My recommendation here is to improve the AEBS.
3. Comments implying that radar would have been better at identifying the ‘cut-off’ are questionable. Lane intrusion is only partial until about 3 seconds before impact. Radar does not return lateral relative-speed, only longitudinal relative-speed. Who knows what lag exists in determining lateral speed and the accuracy of that determination. I doubt that either are very good web based on radar. My guess is that image processing at better than 20Hz would do best in this clear situation.
4. Interpretation of the turn signal can only be done with image processing (to my knowledge.)
5. Nothing is reported about any horn actuation (or if autoPilot even uses the horn). The brake application by the pickup may have been an impulsive response to a horn blow by the Tesla.
6. There seems to be no indication by the driver of the pick-up that he saw the Tesla coming.
7. The Tesla data likely also has its closing speed on the panel truck and thus the closing speed of the pick-up to the panel truck. This information may help us to begin to understand the extent to which the pickup was tailgating the panel truck.
8. To me, AutoPilot’s main issue is: should it allow “passing on the right” when “passing on the right” is illegal. The reason it is illegal is because it leads to crashes like this one, that is an issue that should be taken up by NHTSA and NTSB. To what extent should any of these automated driving devices engage in “illegal” driving? My current view (subject to change) is:
a. Up to 9 mph over is OK.
b. Rolling through a stop sign is OK, if it is determined that time to any likely collision is greater than 5 seconds (meaning you must be able to “see” at least 5 seconds away at speed limit +9 (or something similar)
c. Cross double line as long as oncoming traffic has slowed to under 25 mph and has room to proceed by squeezing right (or something like that).
d. Pass on the right as long as all pertinent vehicles in the two lanes are moving at less than 25 mph (or something like that).
SmartDrivingCars Pod-Cast Episode 222, Zoom-Cast Episode 222
F. Fishkin, July 11, “Is it time for autopilot to not break the law? Princeton’s Alain Kornhauser says yes. And if technology can save lives, prevent injuries and crashes…shouldn’t it? Plus Richard Branson, Jeff Bezos, Waymo, VW and more on Episode 222 of Smart Driving Cars with co-host Fred Fishkin. “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
C. Gohd, July, 11, “Early this morning (July 11), billionaire Richard Branson and three other passengers briefly went to space for the first fully crewed spaceflight of Virgin Galactic’s SpaceShipTwo spaceplane.
At 8:40 a.m. local time (10:40 a.m. EDT; 1440 GMT), the crew of Virgin Galactic’s Unity 22 test flight mission took off from the company’s Spaceport America facility in New Mexico and flew just above the boundary of space, where the four passengers and two pilots experienced about four minutes of weightlessness.
It was “the experience of a lifetime,” Branson said during a live broadcast of the flight. Branson, designated “Astronaut 001″ for the Unity 22 mission, founded the Virgin Group of companies that includes Virgin Galactic. …” Read more Hmmmm… Very impressive from many points of view. Watch video. An Orville & Wilbert moment. Alain
Tesla finally begins shipping ‘Full Self-Driving’ beta version 9 after a long delay: Let the fun begin
A. Hawkins, July 10, Tesla began sending out over-the-air software updates for its long-awaited “Full Self-Driving” beta version 9, the definitely-not-autonomous-but-certainly-advanced driver assist system.
As promised by Elon Musk, the software update (2021.4.18.12) began uploading after midnight on Friday, giving thousands of Tesla owners who have purchased the FSD option access to the feature, which enables drivers to use many of Autopilot’s advanced driver-assist features on local, non-highway streets……” Read more Hmmmm… Unfortunately, the sub-title… “Let the fun begin” is so bad!! FSD should NOT be about “fun”. Maybe “comfort & convenience” but NOT “fun”. This is serious business and the Driver MUST stay completely engaged and ready to take over. In the “1st video” at night with no traffic it didn’t take long before the driver had to intervene. It took less than 2 minutes to have the driver need to intervene to not run over a pedestrian in a cross walk. ” … definitely-not-autonomous … ” End of story!!! Alain
A. Hawkins, July 6, light gray cube with a thin blue top glides down a darkened highway, beset on all sides by dozens of green cubes. The green cubes bounce between lanes in an attempt to pass the gray cube, but the gray cube maintains a steady speed as the blackened landscape slips past into the artificial night.
This is Simulation City, the virtual world where Waymo, an offshoot of Google, tests its autonomous vehicles in preparation for real-world experiences. The gray cube with the blue top represents one of the company’s autonomous semi-trailer trucks, while the green cubes are all the other vehicles on the artificial highway.
Waymo is unique among autonomous vehicle operators in that it has not one but two simulation programs it uses to train its vehicles. The first is CarCraft, which has been in use since at least 2017, and in which Waymo says it has driven over 5 billion miles. Simulation City is the latest virtual world in which the company trains, tests, and validates its “Waymo driver” software in order to ensure its vehicles are better prepared to meet all of the challenges of the open road. Waymo is sharing details about Simulation City for the first time exclusively with The Verge……” Read more Hmmmm…Interesting. See video. Seems to me to be more Simulation Freeway rather than Simulation City. 15 Billion simulated miles seems good; however, if most of those are on Freeways, not so much. (Example: The simulation shown in the article isn’t worth running! However, if the simulations are mostly in the “Trentons” of this world, then I’m impressed.) Alain
F. Lambert, July 10, “….” Read more Hmmmm… Read Lambert’s take and his “1st video” Much the same as above… This is NOT “Full” anything except “snake oil”.
The lane markings are maybe “too good”. They should be extremely good bracketing your lane, the “Blue line” ( where Tesla intends to take you) and the lanes ahead where the blue line intends to take you. Not so much on the lanes next to the blue line lanes. Cross walks and stop lines should also be clear.as well as any lettering that is painted in the lanes (That lettering has been put there to make sure we see it when we are driving.) It is important that Tesla confirms that it sees it too.
Why double yellow lines are not drawn as double yellow lines is disconcerting. It suggests that Tesla designers don’t appreciate the vast amount of thinking and hard work that has gone into the Manual on Uniform Traffic Control Devices (MUTCD) (And, of course, Tesla should incorporate any local variants.)
The purpose of the display is visual information to the alert driver to assure them that the snake oil really understands what’s important about what the Tesla is going to encounter in the next few seconds. The intersection image in this article is very bad at doing that. It is mostly eye candy and not what’s really important ahead. The light is red (It should be bigger). The Tesla must be making a right-turn-on-red (It is in the crosswalk going 25mph (certainly not slow ). No iew is presented showing that the “coast is clear”. Really?? It a camera looking that way. What has it interpreted about what that camera seeing? Does it only display what it sees ahead when it is backing up in reverse? It doesn’t show a cross walk in the lane that it is entering. I bet there is one there. oesn’t it see it? Why is the blue line so short? Does it not know what lane it is intending to go follow? Why all the superfluous other lanes? I could go on but this barely earns a P (Pass) in grade-inflated “Covid-grading”. Alain
R. Bellan, “As environmental issues really came of age in the 1990s, certain German automakers were meeting in secret groups to make sure their cars would continue to industriously contribute to greenhouse gas emissions. According to the European Union, Volkswagen, Audi, Porsche, BMW and Mercedes-Benz parent company Daimler have been illegally colluding to restrict competition in emission cleaning for new diesel passenger cars, essentially slowing the deployment of cleaner emissions tech. On Thursday, the EU issued fines of $1 billion (€875 million) to Volkswagen and BMW for their involvement in the emissions cartel.
“The five car manufacturers Daimler, BMW, Volkswagen, Audi and Porsche possessed the technology to reduce harmful emissions beyond what was legally required under EU emission standards,” said executive VP of the EU Commission Margrethe Vestager in a statement. “But they avoided to compete on using this technology’s full potential to clean better than what is required by law. So today’s decision is about how legitimate technical cooperation went wrong. And we do not tolerate it when companies collude. It is illegal under EU Antitrust rules. Competition and innovation on managing car pollution are essential for Europe to meet our ambitious Green Deal objectives. And this decision shows that we will not hesitate to take action against all forms of cartel conduct putting in jeopardy this goal.”… ” Read more Hmmmm…So very bad!! Shame on them!. We can’t do this kind of thing with SmartDrivingCars. Over-hyping their capabilities comes close. We need to stop that. Alain
S. Alvarez, July 2,”The Tesla Model Y may be stealing some of the Model 3’s luster in the United States, but the all-electric sedan is still a force to be reckoned with in Europe. This became particularly evident in Britain, where the Model 3 became the country’s best-selling car in June, EV or otherwise.
The Model 3’s stellar performance in Britain’s auto market was revealed by the Society of Motor Manufacturers and Traders (SMMT) earlier this week. The SMMT stated that Tesla sold 5,468 Model 3s in June 2021, allowing it to top the country’s list of best-selling cars. The Model 3 also outsold its nearest rival, the internal combustion-powered Volkswagen Golf, by over 800 units. … ” Read more Hmmmm… Congratulations! Alain
S. Crowe, June 21, “t’s officially official. Hyundai Motor Group (Hyundai) announced this morning it completed its acquisition of a controlling stake in Boston Dynamics. Hyundai now owns an 80% stake in Boston Dynamics, while SoftBank owns the remaining 20% through one of its affiliates.
The deal was announced in December 2020. Hyundai paid about $880 million to acquire the controlling stake from Softbank, valuing Boston Dynamics at $1.1 billion….” Read more Hmmmm… Given how much AV companies think that they are worth, $1.1B for Boston Dynamics seems very cheap. By the way… no National Security issues associated with this transaction??? See video. Alain
T Gallen & C. Winston, July 7, “The politics of converting a bipartisan agreement between the White House and centrist senators on a $1.2 trillion infrastructure package into legislation will occupy center stage during the coming months. However, the important economic effects of this package should not be ignored. If passed in its current form, the increased government spending in the legislation may raise U.S. gross national product. But the package will hurt the economy by initiating a costly and lengthy transition to build new taxpayer-funded infrastructure.
Those costs could be reduced, and additional benefits could be realized, if policymakers set efficient user prices to make much better use of the infrastructure we’ve already built. Efficient user prices simultaneously raise a sustainable stream of revenue for expenditures, enable policymakers to reduce expenditures by encouraging users to reduce the costs they impose on other users, and benefit the economy by improving infrastructure performance, such as by reducing congestion and travel delays. The benefits associated with faster and more reliable travel times are generally not included in GNP calculations….
Efficient congestion pricing for cars and trucks and efficient pavement and bridge wear pricing for trucks could reduce the nearly $200 billion in federal and state spending on road infrastructure by as much as one-third, or more than $60 billion. Congestion pricing makes it unnecessary to build additional expensive highway lanes and new roads to accommodate peak-period vehicle traffic, and, as noted, efficient truck pricing reduces maintenance expenditures on roads and bridges. In addition, travelers and shippers benefit from faster and more reliable trips and from smoother pavement that reduces vehicle repair costs.” Read more Hmmmm… All back to Wm. Vickery who earned a Nobel for it. By the way, it should be called “Value Pricing” (It is all about adding “Value” to those involved, Plus it sounds positiv instead of “Congestion Pricing” (Sounds so negative). Alain
T Gallen & C. Winston, July 2021, “We analyze the effect of the US transportation system on economic activity by building a quantitative dynamic general equilibrium model with a taxpayer-funded transportation capital stock. We highlight stark differences between the positive welfare effects of additional infrastructure spending in the long run, and its potentially negative effects when we account for the large transition (time and delay) costs to build. We also quantify large differences between the effects of additional infrastructure spending and efficient transportation policies, such as congestion pricing and eliminating laws that artificially inflate input prices, concluding that taxpayer-funded transportation improvements that increase GDP significantly may produce smaller welfare gains than efficient policies that increase GDP modestly…” Read more Hmmmm… Details in support of the above. Alain
A. Hawkins, July 10, “Tesla CEO Elon Musk is finally admitting that he underestimated how difficult it is to develop a safe and reliable self-driving car. To which the entire engineering community rose up as one to say, “No duh.”
Or at least that’s how it should have happened in a just world. Instead, all the Tesla sycophants and ass-kissers on Twitter told Musk to keep up the good work, that they believed in him, and encouraged him to hurry up and roll out the latest version of his “Full Self-Driving” software that, it’s worth pointing out, does not enable a Tesla vehicle to drive itself without input from the driver……” Read more Hmmmm… OK, he might have moved a couple of degrees, but he needs to do a complete 180, a Full Monty. Alain
Re-see: Pop Up Metro USA Intro 09 2020
K. Pyle, April 18, “â€œItâ€™s time to hit the start button,â€ is Fred Fishkinâ€™s succinct way of summarizing the next steps in the Smart Driving Car journey. Fiskin, along with the LA Timesâ€™ Russ Mitchell co-produced the final session of the 2021 Smart Driving Car Summit, Making It Happen â€“ 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 â€œthe Trentons of the World.â€
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“
C’mon Man! (These folks didn’t get/read the memo)
Calendar of Upcoming Events
Virtual on July 12-15, 2021
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