Sunday, January 12, 2020
SmartDrivingCar.com/8.02-USDoT_v4-011220
2nd edition of the 8th year of SmartDrivingCars
Hmmmm… Reflections
A. Kornhauser, Jan 12, Hmmmm… Self-driving cars are hot and the OEMs are responding. I’m about to buy a new Subaru Outback and EyeSight is standard. It is no longer just AutoPilot or expensive options that car salesmen don’t sell. Car companies, as reflected in what is in showrooms and what was promoted at CES, have realized the comfort and convenience of Self-driving technology (cars that have a lot of the Safe-driving car features but also enable you to take your feet off the pedals and hands off the wheel at least for short periods of time. These technologies are really becoming the ‘chrome and fins’ that sell cars to individuals in the 2020s. The momentum is all behind that happening and there is little Washington or Trenton or Princeton Council can do about it. Hopefully part of that momentum will be to make these systems actually work well, especially the Automated Emergency Braking Systems (MUST quit assuming that all stationary objects in the lane ahead can be passed under and consequently each is disregarded. As Tesla is finding out, sometimes those objects are parked firetrucks.) and begin to put hard limits on over-speeding, tailgating and use while driver is impaired. Self-driving cars are unfortunately going to lead to substantial urban sprawl, increased VMT, increased congestion and do nothing to help the energy and pollution challenges of our addiction to the personal automobile. Only ‘Waymo-style Driverless’ (autonomousTaxis, (aTaxis)) tuned to entice ride-sharing can potentially stem the tide of ever more personal car ownership and ever expanding urban sprawl. Alain
Smart Driving Cars Podcast Episode 137
F. Fishkin, Jan 12, "The Transportation Department’s update on autonomous vehicles. What’s missing? Princeton’s Alain Kornhauser joins co-host Fred Fishkin for that plus the latest from Mobileye, Qualcomm, Aptiv, Tesla and more. " Just say "Alexa, play the Smart Driving Cars podcast!". Ditto with Siri, and GooglePlay … Alain
Ensuring American Leadership in Automated Vehicle Technologies: Automated Vehicles 4.0
NSTC & US DoT, January 2020. "The United States Government is committed to fostering surface transportation innovations to ensure the United States leads the world in automated vehicle (AV) technology development and integration while prioritizing safety, security, and privacy and safeguarding the freedoms enjoyed by Americans. The U.S. Government recognizes the value of industry leadership in the research, development, and integration of AV innovations. Such innovation requires appropriate oversight by the Government to ensure safety, open markets, allocation of scarce public resources, and protection of the public interest. Realizing the full potential of AVs will require collaboration and information sharing among stakeholders from industry, State, local, tribal, and territorial governments, academia, not-for-profit organizations, standards development organizations (SDO), and the Federal Government.
AV 4.0 presents a unifying posture to inform collaborative efforts in automated vehicles for all stakeholders and outlines past and current Federal Government efforts to address these areas of concern. AV 4.0 establishes U.S. Government principles that consist of three core interests, each of which is comprised of several sub-areas…." Read more Hmmmm…short on substance, except appropriate emphasis on safety (absolutely necessary) and a weak mention of mobility (4th place on page 1. Mobility should be in 2nd place.). The focus remains on automation that enhances the personal automobile as opposed to how this technology could provide mobility that is so good it could replace the personal automobile. What is not addressed is that enhancing the personal automobile has substantial negative societal implications. (VMT, congestion, discrimination,…)
In the short run, automation aimed at overcoming mis-behavior by drivers in the form of Safe-driving cars is fundamentally good as is technology that allows us to sometimes take our hands off the wheel and feet off the pedals in Self-driving cars. This document encourages and supports the auto industry to do as much as it can with these technologies. It doesn’t go the full distance by endorsing automated technologies that prohibit excessive speeding, tailgating, running of red lights and impaired driving, each of which are easily doable, but might engender enormous angst by an industry that has traditionally allowed and implicitly promoted the blatant mis-use of its vehicles by its customers. It also continues to promote the concept of personal Driverless cars without the caution of the responsibilities that would necessarily need to be imposed on individuals as they dispatched these vehicles to go get stuff for them. The problem is not when a responsible owner is inside the vehicle if something bad happens. But, what is the deal when the owner/dispatcher is nowhere to be found? This doesn’t seem to be a vision that we should be promoting. In fact I think that ownership/control of driverless vehicles by individual consumers should be prohibited.
As mentioned above, automation as promoted in this document will exacerbate the sprawl, congestion, energy, pollution and VMT aspects of our dominant form of mobility. While safety will surely be enhanced all other societal objectives will suffer. Hopefully version 5 will focus on enhancing PMT (person miles traveled) while reducing VMT (vehicle miles traveled)
In the longer run, this technology, in the form of driverless mobility machines, may be able able to deliver desirable mobility that for many may entice them to eave their cars at home. Uber/Lyft have given some of us a taste of what such mobility might look like if it is done really well. We always hailed cabs in NYC and if we were really desperate/stuck called them in the suburbs. But Uber/Lyft made it so easy and largely anxiety free that I and many others no longer drive and park our own cars, nor rent a car, when we fly somewhere. We simply use Uber/Lyft. Price doesn’t really matte since it is often someone else that is paying. The comfort, convenience and anxiety-free aspects of Uber/Lyft win out. For these trips, Uber/Lyft are just fine; they are largely price inelastic. No need to replace the driver.
However, for most other trips, mode choice (aTaxi or personal car) is very price/perceived-cost sensitive, especially when considering the marginal cost of using one’s personal automobile and perceiving that the cost of driving it yourself is absolutely zero. For these trips, automated technology makes the cost of being chauffeured essentially free. It transforms the Uber/Lyt mobility model to such a large extent that enables it to become THE mobility of choice for many more trips. Eliminated is the lingering anxiety associated with the gig driver, plus its cost/fare price is competitive to the perceived cost of drive yourself. For many, it will be more affordable than using a personal car for many if not most trips. It will be available for many who today can’t or choose not to drive. Plus, opportunities for ridesharing can be promoted that address many of the societal challenges of the personal automobile. Such a vision is essentially absent from this document.
There is a section on "Review traffic laws and regulations that may serve as barriers to operation of ADSs" (p21) that is woefully inadequate. The big issues with traffic laws have to do with enforcement. If the speed limit is X is its enforcement going to be different if an automated system is controlling the vehicle as opposed to a human driver? Is the Driverless vehicle operator included in the morning police briefing when it is decided "if they go Y or more above X, pull them over and ticket them, else let them go?" This section doesn’t even acknowledge that such a fundamental issue exist. Alain
Feds tweak driverless-car guidelines, seek to balance safety and tech development
R. Mitchell, Jan. 8, "Driverless cars and trucks will be hitting the highways in increasing numbers over the next few years. The U.S. Department of Transportation doesn’t want to get in the way.
That’s the message in a new set of guidelines the department released Wednesday. The intent is to spur further development while emphasizing safety, the department said.
“Safety is always No. 1 at the U.S. Department of Transportation,” department Secretary Elaine Chao said in a short speech at CES, the big consumer electronics show in Las Vegas, where the new guidelines were announced. But “remaining technology neutral” is a department commitment and “protecting American innovation and creativity” is another top priority…
The biggest change in the new set of guidelines, called Automated Vehicles 4.0, is a streamlined system of federal oversight. Without offering specifics, Chao said the new guidelines “unified AV efforts across 38 federal departments, independent agencies, commissions and executive offices of the president.”.." Read more Hmmmm… A good read. Alain
Waymo is way, way ahead on testing miles—that might not be a good thing
T. Lee, Jan. 7, "Waymo, Alphabet’s self-driving company, has logged 20 million miles on public roads, the company announced in a Tuesday press statement. The new milestone comes just 15 months after Waymo hit the 10 million mile mark in October 2018. The latest figure puts Waymo far, far ahead of its rivals….
The company has been making slow and steady progress. Waymo says it has completed some fully driverless rides for passengers in its closed testing program in a portion of the Phoenix area. But it’s not clear how soon this fully driverless technology will become available commercially in Phoenix. And it’s even less clear how quickly Waymo will be able to scale up to other cities and eventually the world. What if Waymo’s approach is a dead end?…" Read more Hmmmm… Very interesting, but a dead end is very unlikely, simply because there is enormous latent demand for affordable mobility-as-a-service rather than ‘I’ve got to do it all myself (buy a car, drive it, insure it, maintain it, park it, clean it, keep it from being stolen, …)’. Waymo has the opportunity to serve that fundamental consumer demand by simply putting what they’ve learned in all of that testing on a commodity (existing cars/Chrysler Pacificas) using technology that obeys Moore’s Law and a physical infrastructure (our existing roads and bridges) that is desperately in need of a mobility system that more effectively and responsibly uses what exists (ride sharing of vehicles and elimination of mis-behavior). They are better poised to successfully deliver safe mobility in an Operational Design Domain (ODD) that has real customers and can more readily grow rather than low speed shuttles that have essentially no opportunity to grow beyond their niche application. 25 years ago we built the Morgantown PRT system which was the low-speed automated shuttle of its day. It has been enormously safe and successful; however, it was so niche that it was never expanded. It never scaled. It never came close to scaling.
Waymo’s approach (and that of GM/cruise and Ford/Argo) is one that creates a initial niche (ODD) that has the opportunity to not only be successful, but also is poised to scale and grow exponentially. If that is a dead-end, then they will, and should, fail. I don’t think that is a dead end. 25 mph shuttles are nice in the ‘Morgantowns’ of this world, and face enormous hurdles if they wish to become fundamental staples of mobility. Waymo has the opportunity and is shooting to become a fundamental staple of mobility. Alain
Intel’s Mobileye has a plan to dominate self-driving—and it might work
T. Lee, Jan. 10, "A lot of media coverage of self-driving technology has focused on a handful of big companies with well-known brands: Google, Uber, Tesla, and GM. But there’s another company working on self-driving technology that might ultimately prove even more important. That company is Mobileye, an Israeli startup that was acquired by Intel in 2017.
Mobileye doesn’t have Elon Musk’s star power or Google’s billions. But it has something that’s arguably even more important: a dominant position in today’s market for advanced driver-assistance systems (ADAS). Mobileye had a very public split with Tesla back in 2016, but it continues to do business with a lot of other carmakers. Mobileye says it shipped 17.4 million systems last year, which means 17.4 million customers bought cars with Mobileye’s cameras, chips, and software…." Read more Hmmmm… This is all about ‘Self-driving’ and NOT ‘Driverless’ mobility. In fact, it is the competition to Driverless. Its intent is not to provide mobility-as-a-service, but instead continue to induce all of us to furnish our own mobility. This is the ‘chrome & fins’ that will propel the ‘own-your-own’ car industry through the 2020s. Don’t be surprised if that industry begins to throw monkey wrenches and Molotov cocktails into the "Waymo-like" driverless cars initiatives. Alain
Tesla is challenging hackers to crack its car, and it is putting ~$1 million on the line
F. Lambert, Jan. 10, "…Tesla has been investing more in its cybersecurity over the last few years, and now it is returning to Pwn2Own to challenge hackers to crack its cars with ~$1 million on the line and a few Model 3 vehicles.
Last year, Tesla went to Vancouver for Pwn2Own, which is a hacking competition run by Trend Micro’s Zero Day Initiative (ZDI). It offered a Model 3 to whoever manages to find and exploit certain vulnerabilities in the vehicle’s system. A hacking duo targeted the infotainment system on the Tesla Model 3 and used “a JIT bug in the renderer” to manage to take control of the system. They left with a brand new Model 3.
These types of hacking competitions with white-hat hackers enable Tesla to test and improve its security systems, which are becoming increasingly important in cars as they are becoming more like computers on wheels…
This year, they are making ithe Tesla challenge a lot more complex:..’ Read more Hmmmm… Brilliant and cost effective. Alain
Qualcomm Moves Toward Self-Driving Vehicle Tech
TM. Freeman, Jan. 9, "… Qualcomm announced Snapdragon Ride on Jan. 6 at CES, the annual technology trade show in Las Vegas. Its foray into safety-autonomy systems pits it against several large rivals, including Google-parent Alphabet-Waymo and Intel-Mobileye…Qualcomm is betting that the Snapdragon Ride platform can reduce the complexity of current driver assistance and autonomous driving technology with a comprehensive software suite, power-efficient hardware, computer vision and other techniques to power such things as automatic braking and traffic sign recognition.
Snapdragon Ride includes processor and accelerator chips, artificial intelligence and sensor fusion technologies, as well as what the company called a pioneering “software stack.” It is designed to be power-efficient, which is important as electric vehicles become more widespread…" Read more Hmmmm…See video Alain
Tesla driver charged with careless driving for flossing his teeth on Autopilot at 84 mph
F. Lambert, Jan. 10, "…If the risk of an accident is not enough to deter you from not paying attention when driving on Autopilot, hopefully the risk of being cited for reckless or careless driving might help..’ Read more Hmmmm… Well said. (Is the speed limit in Ontario above 80mph? Why didn’t he also get a speeding ticket?) Alain
Self-Driving Cars May Never Dominate…And That’s OK
S. Corwin, Jan. 10, "…Maybe AVs have been an illusive shiny object that has been over-hyped as a barometer of the future of mobility. But that’s not to say autonomous vehicles won’t be on our streets soon, or that they will necessarily have a deleterious impact. Like all technologies, they are simply a tool. Whether they further or hinder our broader goals as a society hinges on how we collectively choose to integrate them into the urban fabric. …." Read more Hmmmm… Interesting. Alain
Tesla (TSLA) rallies near $500 per share as shorts go mad
F. Lambert, Jan. 8, "…A potential short squeeze could also be helping Tesla’s stock as some traders with short positions on Tesla might still be covering their positions. The Tesla short community is generally pretty active on Twitter under the tag $TSLAQ, but they have been a lot quieter during this rally. Many of those who are still commenting are going mad:..’ Read more Hmmmm… Life is good. Disclosure: I don’t own TSLA :'( Alain
McCarran International Airport tests self-driving cars
K. Galer, Jan. 10, "…It’s the first airport to allow the autonomous cars, but it’s still in the testing phase while Aptiv works out the kinks…. They’re easy to spot. Aptiv has a fleet of black and white BMWs with orange wheels. They’ll take passengers to more than 3,400 destinations in the valley. …. " Read more Hmmmm… More working out of the kinks in real environments. Alain
C’mon Man! (These folks didn’t get/read the memo)
Half-Baked
Feds Are Content to Let Cars Drive, and Regulate, Themselves
A. Marshll, Jan. 9, "A new Transportation Department policy on self-driving cars is long on boosting the industry and short on ensuring its safety." Read more Hmmmm…Actually it is very long on Safety, DoT’s primary reason for existence. It is very short on everything else. The public sector is completely out matched compared to the private sector on everything else having to do with automation. Alain
Corrections
Calendar of Upcoming Events:
4th Annual Princeton SmartDrivingCar Summit
evening May 19 through May 21, 2020
Princeton University
Princeton, NJ
On the More Technical Side
https://orfe.princeton.edu/~alaink/SmartDrivingCars/Papers/