Monday, June 28, 2021
24th edition of the 9th year of SmartDrivingCars eLetter
How Important is Exact Localization for HAD?
M. Sena,July/Aug. ’21, “In this issue of The Dispatcher for July and August, I have taken up a subject in the lead article that has been on my list for quite some time. It is of how cars that drive themselves keep themselves on the road while they make their journey to their destination. It turns out that there is a very good reason why Teslas crash and Waymo is running around only in Chandler, Arizona after people who didn’t know better promised that there would be completely driverless cars on all roads a decade ago: localization of a moving vehicle is very, very hard, even for a human.
I encourage you to read Musings this month. It’s about making the journey to a world without climate change protests, a world where they either won’t be necessary or allowed. On most journeys, we have to cross bridges. Sometimes we have to make them ourselves. Think of the article as the first bridge to cross toward a better understanding of the climate change journey.
Dispatch Central contains, as usual, something for everyone. Insurance is addressed in the two main articles. In Bits and Pieces I have added my thoughts on recent events.
This is a double issue, in part because we are going to try to do more this summer than we could do last. But it’s also because I need some extra time to work on a follow-up to the Princeton SmartDrivingCars Summit with Professor Alain Kornhauser. There was a concrete proposal put forward by Professor Kornhauser during the last session, and many of us who took part in the Summit have committed to try to work on implementing that proposal. Read more Hmmmm… . Once again an outstanding The Dispatcher. I happen to have a diffent fundamental view on”exact localization” than Michael, many and possibly even everyone else… As usual, I’ll take a very self-centered view… I’ve lived my whole life without knowing (or caring to know) my “exact location”. I’ve been satisfied to know: “sort of… where am I?” but exact…where am I? … not so much. What troubles me about the “exact where am I” is that this exactness is in some coordinate system. Where is the origin of that coordinate system and is moving? Oh, it’s the “center” of the earth?? Or some “reference point”. So “exact” is actually, “exact relative to some reference point. Little seems to ever be said about the “exactness” of the reference point, but that may actually be some saving grace about “exact”.. it is “exact” relative to some reference point.
I see.. If the reference point is the center of the Universe, then I’d better be really-really precise; else, small small changes mean big-big differences. If the reference is the center of the earth, then I may just need to be really precise; else, small changes mean big differences. However, if the reference point is my nose and I’m trying to stay between two white lines and not hit anything, then the precision to which I need to know where things are may not need to be very precise as long as I have a little bit of leeway and still stay between the lines and leave enough room around the various objects to not hit them.
OK, safe driving requires only knowing where I am relative to objects around me to a moderate level of precision. I can do it in two ways… take the difference between two values: location of object and my location. The farther away the reference point, the more precise they will need to be if precision of the difference is to be maintained. Consequently, if the measurements are relative to my nose, the need for about as small as it can get.
Moreover, any precision data base lacks some “most” important values.. 1. a precise value for my location and 2. a precise value for anything around me that moves (meaning it wasn’t at its current location when the HD database was assembled). Required is the ability in real time to locate and track objects relative to me (my nose, the hood ornament of my car, ..) with only some precision These objects and their location aren’t included in these precise/HD databases. What is needed is a very reliable means of identifying objects and determining their position and velocity with little latency. This is absolutely necessary;uyr;y necessary for the moving objects, might as well do it also for the stationary objects. 😁 Alain
Please don’t suggest that one needs an HD map database in order to run their SLAM (Simultaneous Localization And Mapping) algorithm. That algorithm needs as input the relative position (sensor observations) of objects . The capability to determine those inputs is all that is needed to do collision avoidance, so don’t even bother going through the SLAM computation and certainly don’t pay for a reference data set.
SmartDrivingCars Pod-Cast Episode 219, Zoom-Cast Episode 219 w/Michael Sena, Editor, The Dispatcher
F. Fishkin, June 29 , “Why couldn’t a smart driving car prevent Alain’s crash with a deer? How important is exact location for highly automated driving? And NHTSA wants reports on all automated vehicle system crashes. The Dispatcher publisher Michael Sena joins Princeton’s Alain Kornhauser and co-host Fred Fishkin for that and more. “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
Are self-driving cars safe? Highway regulator orders industry to cough up the data
R. Mitchell, June 29, “After years of inaction, the federal government will begin collecting crash data on automated vehicles.
The National Highway Traffic Safety Administration on Tuesday ordered dozens of car, truck and technology companies to inform the agency of a serious crash within a day of learning about it, with a more complete data report due after 10 days.
The order will enable NHTSA to “collect information necessary for the agency to play its role in keeping Americans safe on the roadways, even as the technology deployed on the nation’s roads continues to evolve,” the agency said.
The order applies to highly automated vehicles, including robotic cars that don’t require a human driver, as well as partially automated systems such as Tesla’s Autopilot and General Motors’ Super Cruise with advanced cruise control and automatic steering.
It immediately affects the partially automated so-called Level 2 systems increasingly common on new vehicles from most major manufacturers. The number of fully robotic cars and trucks now deployed on public roads is tiny, but the market is expected to grow dramatically in coming years.
Manufacturers tout the safety and convenience of automated vehicles, but scant useful data have been collected to demonstrate how safe they are.
“This is very important. It’s fantastic. And it’s about time,” said Alain Kornhauser, who heads the automated vehicle engineering program at Princeton University. “Safety should not be a competition. It’s a cooperation.”…
“Nobody should push back on this,” Princeton’s Kornhauser said. “We don’t know what we don’t know, we don’t know what works and doesn’t work, and this allows us to begin to know that.”…” Read more Hmmmm… I couldn’t have said it better myself. 😁 Alain
NHTSA Orders Crash Reporting for Vehicles Equipped with Advanced Driver Assistance Systems and Automated Driving Systems
Staff, June 29,” The National Highway Traffic Safety Administration (NHTSA) today exercised its authority by issuing a Standing General Order requiring manufacturers and operators of vehicles equipped with SAE Level 2 advanced driver assistance systems (ADAS) or SAE Levels 3-5 automated driving systems (ADS) to report crashes. This action will enable NHTSA to collect information necessary for the agency to play its role in keeping Americans safe on the roadways, even as the technology deployed on the nation’s roads continues to evolve. …” Read more Hmmmm… Excellent. You MUST read the Standing General Order. Alain
A new bar for safety
M. Rosekind, June 2021,”…Our design objective is for our vehicles to be free of any single point of failure from a safety perspective. So we’ve designed a number of innovative and redundant features that allow the vehicles to continue to operate safely even if certain safety systems become unavailable. We rigorously test and validate our autonomous driving technology to ensure safe operations within our operational design domain (ODD). Before tires ever hit pavement, we test our autonomous technology via simulation and an integrated hardware/software system affectionately called LabBot….” Read more Hmmmm… . There is a lot here. A must read as are each of the driverless vehicle developer’s Safety reports. Alain
How to build a driverless vehicle that doesn’t make you barf
A. Hawkins, June 22, “Zoox released the second version of its safety report today, containing new details about the vehicle that the Amazon-owned company is building from the ground up to be completely autonomous. The vehicle, which looks like an oversized toaster on wheels, lacks typical controls like a steering wheel and pedals and is designed to be bidirectional, meaning it can travel in either direction. And while that could in theory increase the chances of passengers getting sick, Zoox says its specialized design will make it one of the smoothest rides around..
Zoox is also touting its vehicle’s bidirectional capabilities, meaning it will never have to “back up,” which should improve passenger pickups. “No more U-turns, no more three-point turns,”…”Read more Hmmmm… Agreed that bi-directional capability is important; however, it is hard to imagine that the positives outweigh the negatives. Directionality is important in travel of people. Packages, are indifferent, so may be OK for package delivery in the early morning hours. Up and readily back out of a driveway and keep going. Alain.
Amazon Snaps Up Option to Buy Stake in AI Truck-Driving Startup
L. Chen, June 21, “Amazon.com Inc. has placed an order for 1,000 autonomous driving systems from self-driving truck technology startup Plus and has acquired the option to buy a stake of as much as 20%, Plus said in a regulatory filing, confirming an earlier Bloomberg report.
Amazon has the right to buy preferred shares of Plus via a warrant at a price of $0.46647 per share, the filing shows. That amounts to a roughly 20% stake based on Plus’s shares outstanding before its planned merger with special purpose acquisition company Hennessy Capital Investment Corp. V.
The Sequoia Capital China-backed company, which is developing autonomous driving technology for long-haul trucking, is set to have a valuation of $3.3 billion, adding $500 million in proceeds to accelerate its expansion, the company said in a statement in May. The company raised $150 million via so-called private investment in public equity, or PIPE, from funds including BlackRock Inc. and D.E. Shaw….” Read more Hmmmm… Maybe? Here is Henessy’s Registration Statement from Friday, June 28. Another SPAC in this space. Got a bump on June 18, 10 days prior to the registration statement. How does that happen???? Is Plus really better than Waymo, Aurora, Zoox, Robotic Research, TuSimple, … No FOMO here. Alain
Amazon eyes robot truck startup as it continues to hedge its bets on AV technology
A. Hawkins, June 21, “Amazon owns a robotaxi company, is experimenting with delivery robots that drive on the sidewalk, and now is considering acquiring a big stake in a robot trucking startup.
According to Bloomberg, the e-commerce giant recently placed an order for 1,000 autonomous driving systems from Plus, a California-based company working on driverless trucking technology. Amazon is also considering buying a 20 percent stake in Plus — a move that could have serious implications for how the company moves its products around the world.
Amazon is thinking about buying as much as 20 percent of Plus, which recently announced plans to go public via a reverse merger with a special purpose acquisition company, or SPAC. The company plans to merge with Hennessy Capital V, a SPAC that also sponsored the public debut of EV startup (and recent target of a Securities and Exchange Commission investigation) Canoo….” Read more Hmmmm… Andrew’s take on Amazon’s initiative. Alain
Volvo won’t charge extra for its upcoming electric vehicles with lidar sensors
A. Hawkins, June 21, “Volvo plans on offering the lidar — and artificial intelligence-powered safety features — of its upcoming electric vehicle as standard, but the Swedish automaker plans on charging extra for its semi-autonomous systems, like the hands-free Highway Pilot.
Volvo’s fully electric successor to the XC90 SUV, to be revealed in 2022, will have two new components that have never appeared on a Volvo vehicle before: a lidar sensor produced by Luminar and an onboard “autonomous driving computer” powered by video game hardware manufacturer NVIDIA.
Those two pieces of hardware will help enhance Volvo’s traditional safety features, like automatic emergency braking and blind spot detection, before the company is ready to start rolling out more autonomous features to customers, said Volvo chief technology officer Henrik Green.”….” Read more Hmmmm… Hopefully the LiDar will make essentially perfect Volvo’s Automated Emergency Braking system. No more rear ending, deer strike or any other collisions. I hope. Can this become true? Safety STARTS with an Automated Emergency Braking System that actually prevents collisions. If these LiDar sensors don’t do that, then they should be free because they aren’t worth anything. Alain
The Boring Company Begins Operations at Las Vegas Loop
S. Levine, June 9, “For people reluctant to gamble with waiting in Las Vegas traffic, The Boring Company has a new option: going underground. Elon Musk’s subterranean venture announced that operations at its Las Vegas Loop have begun this week.
The $48.7 million project enables people to visit one of three stations and hitch a ride in a Tesla (NASDAQ:TSLA) Model 3 or Model X to travel around the 1.5 mile loop that connects the Las Vegas Convention Center campus. Based on testing that the company completed in May, the Loop can transport more than 4,400 people hourly at a speed of about 35 miles per hour — a far cry from the autonomous travel at high speed that was the original vision of the project….” Read more Hmmmm… It is a start. Now let’s see if it can blossom. Alain
Luminar is using lidar to help build the ‘uncrashable car’
A. Hawkins, June 15, “…Lidar, a key ingredient in autonomous driving, is a laser sensor that uses near-infrared light to detect the shapes of objects. This helps autonomous vehicles “see” other road users like cars, pedestrians, and cyclists, all without the help of GPS or a network connection…..” Read more Hmmmm… Please… Forget about autonomous anything until you finally get Automated Emergency Braking to work “perfectly”. “Seeing” other road users doesn’t “need” GPS nor have time for a “network connection” The above statement is just noise. Agan.. a Necessary Condition to have an uncrashable car is and Automated Emergency Braking System that is essentially perfect. It requires the real-time determination of the “free-volume” ahead such that the car does not crash into anything. That free-volume is not only “width” and “depth” but also “height” . Just ask Joshua Brown. Alain
Re-see: Pop Up Metro USA Intro 09 2020
H. Poser’77, Sept 13, 2020. “Creating Value for Light Density Urban Rail Lines” . See slides, See video Hmmmm… Simply Brilliant. Alain
4th Annual Princeton SmartDrivingCar Summit It is over!!! Now time to actually do something in the Trentons of this world.
Making Driverless Happen â€“ The Road Forward (Updated)
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
14th Session What Will Power Safely-driven Cars
13th Session Improving the Moving of Goods
12th Session 3/18/21 Human-centered Design of Safe and Affordable Driverless Mobility
11th Session 3/11/21 Incentivizing Through Regulation
10th Session 3/04/21 Incentivizing Through Insurance
9th Session 2/25/21 Can Level 3 be Delivered?
8th Session 2/18/21 Who Will Build, Sell and Maintain Driverless Cars?
Michael Sena’s Slides, Glenn Mercer Slides
7th Session 2/11/21 Finally Doing It
6th Session 2/ 4/21 Safe Enough in the Operational Design Domain
5th Session 1/28/21 At the Tipping Point
4th Session 1/21/21 Why Customers are Buying Them
3rd Session 1/14/21 The SmartDrivingCars We Can Buy Today
2nd Session 1/ 7/21 A Look into the Future
1st Session: 12/17/20 Setting the Stage
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
The 2021 TRB Annual
Automated Road Transportation Symposium
Virtual on July 12-15, 2021
5th Annual Princeton SmartDrivingCar Summit
Live in Person
To be Announced
On the More Technical Side
K. Lockean’s AV Research Group at U of Texas
and The SYMPOSIUM ON THE FUTURE NETWORKED CAR 2021 VIRTUAL EVENT
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