Earlier this week, the U.S. House Subcommittee on Digital Commerce and Consumer Protection held a hearing on Self-Driving Cars: Road to Deployment. I know, it sounds super boring, and most of it was: if you’ve been following the space for a while, nothing in the prepared statements will surprise you all that much, even though the witnesses at the hearing included industry heavy hitters like Gill Pratt from TRI, GM’s Vice President of Global Strategy Mike Ableson, and Anders Karrberg, Vice President of Government Affairs at Volvo Car Group, as well as Lyft’s Vice President of Public Policy Joseph Okpaku and Nidhi Kalra, Co-Director and Senior Information Scientist, at the RAND Center for Decision Making Under Uncertainty.
What was interesting, however, was the question and answer session. It’s an open look at what the house members think is important, and the answers from the witnesses are on the fly. Remember, these are the people who are making self-driving car policy talking directly to the people who are making self-driving cars: What was talked about at this hearing could potentially shape the direction of the entire industry. There’s over an hour of questioning, and we’ve watched it all. But we opted to transcribe only the most interesting bits.
If you’d prefer to subject yourself to the entire Q&A session, it starts here. Again, note that we’ve just excerpted what seems most interesting to us, so this is not a complete transcript.
Frank Pallone (D-N.J.): Volvo has said that it will skip Level 3 automation and go from Level 2 to Level 4. Can you explain that decision?
Anders Karrberg, Volvo: At Level 3, the car is doing the driving. The car is doing the monitoring. But, the driver is the fallback. So, you could end up in situations where the driver has to take back control, and that could happen within seconds. We are concerned about the Level 3 stage, and therefore we are targeting Level 4.
Nidhi Kalra, RAND: I agree. There is evidence to suggest that Level 3 may show an increase in traffic crashes, and so it is defensible and plausible for automakers to skip Level 3. I don't think there’s enough evidence to suggest that it be prohibited at this time, but it does post safety concerns that a lot of automakers are recognizing and trying to avoid.
Pallone: Volvo has said that it will take complete liability at level 4. Can you explain that decision?
Karrberg: Car makers should take liability for any system in the car. So, we have declared that if there’s a function to the [autonomous driving] system when operating autonomously, we would take the product liability.
Pallone: How real is the threat of hacking, especially in the autonomous context?
Gill Pratt, TRI: I think it’s important to understand that as serious as this threat is, there are also mitigations that we can employ. First of all, we can make sure that the safety technology on the car does not depend on the wireless network in order to operate. Our philosophy is that all the safety functions have to be self-sufficient on the car itself.
Gregg Harper (R-Miss.): Can you elaborate on the work that [your companies] are doing to provide transportation and access [for people with disabilities] in the future?
Mike Ableson, GM: It’s a very exciting opportunity for these communities. While we recognize the potential benefits, there’s obviously a whole lot more work that needs to be done. However, inside General Motors we have a specifically designated employee resource group composed of people with various physical challenges, and they’re already working with our engineering group on the potential for self-driving vehicles going forward.
Pratt: Our president, Akio Toyoda, decided to change the company policy on autonomous driving as the result of a meeting with a blind person, who asked him, “Can I enjoy the mobility of your cars as well?” I wanted to add one more thing: we can’t forget about aging society. Right now, in the United States, 13 percent of our population is over age 65. Because of the baby boom, in 15 years, that fraction will grow from 13 percent to 20 percent, and this is an extraordinary thing. My sister and I had the experience of having to take away the car keys from my father, because he was too elderly to drive. That’s something I don’t think anyone should have to go through. Our goal is to make that not have to happen in the future.
Karrberg: We fully recognize the potential for safe driving cars to bring a happier life to disabled people. Every Sunday I meet my father, who just turned 100, and he asks me every time: When can I have this car? For Volvo initially, we are targeting commuting, because that’s where we think the biggest benefit and interest for consumers are.
Joseph Okpaku, Lyft: We’ve already heard from the disabled community about how much ridesharing has increased their quality of life and mobility, same thing for the senior population. In terms of the potential to have that same impact with autonomous vehicles, the role that ridesharing plays is the ability to bring AVs to the market at a scale that would address this issue in a broad and sweeping way.
Debbie Dingell (D-Mich.): Are there specific things that congress should avoid doing that would stifle the development of autonomous vehicles?
Ableson: We wouldn’t want to see [the] government taking steps to specify a specific technology or specific solution. I think as long as we keep in mind that the goal is to prove that the vehicles are safer than drivers today, the NHTSA guidelines published last year are a very good step in that direction, in that they specify what the expectations are before vehicles are deployed in a driverless fashion.
Pratt: An evidence based approach is really the best one, where the government sets what the criteria are for performance at the federal level, but does not dictate what the ways are to meet that particular level of performance.
Brett Guthrie (R-Ky.): Do self-driving cars have to be perfect to allow them on [the road], and how do we get to the point where they’re safe enough?
Ableson: I think the point is that there’s no way to prove statistically that something’s perfect. We have to agree on the metrics by which going to use to show that the vehicle is better than human drivers, and is therefore appropriate to start deploying without drivers. That’s why this testing in real world is so important, because you’ll see those real-life conditions that we all deal with on a daily basis as human drivers, and we’ll make sure that the vehicles can react appropriately.
Pratt: This is a question that we’re thinking about extremely deeply now. We feel that there may need to be a safety factor multiplying human performance. In other words, if an autonomous car is only slightly better than the average human driver, that may not be good enough, because emotionally, we can empathize with a human driver that has an accident because that could have happened to us. On the other hand, when a machine makes a mistake, our empathy is much less. We don’t know what the safety factor has to be, and what we would like is to work collaboratively with government to try to figure out what that answer is. We’re worried that it may not be one.
Doris Matsui (D-Calif.): Can you provide your perspective on where regulation might be needed at both the state and federal levels?
Pratt: It’s the federal government that we believe should take the leading role. As you may know, in California there’s a requirement, if you’re doing autonomous car development, that you report to the government what your disconnection rate is—every time that your car has a failure of a certain kind. That’s not such a bad idea, but that information then becomes publically available, and it creates a perverse incentive, and the incentive is for companies to try to make that figure look good, because the public is watching. And that perverse incentive then causes the company to not try to test the difficult cases, but to test the easy cases, to make their score look good. We think it’s very important that there be deep thought about this kind of issue before these rules are made, and I think that concentrating that thought in the federal government is the best idea.
David McKinley (R-W.Va.): What is the projected additional cost per vehicle?
Pratt: The costs presently are very high, in the many thousands if not tens of thousands of dollars. Part of the reason that you’re seeing a push to use it in rideshare systems at the beginning is because there you can amortize the cost over a higher utilization of rideshare vehicles. However, we should keep in mind the incredible rate of decreasing costs in the electronics industry, particularly with scale. Think about your cellphone, and the cost of the camera that’s inside your cellphone, which rivals some of the best cameras that you could buy for personal or professional use in the past, and these now cost pennies to put inside of a cellphone. So we don’t know the actual numbers, but we are confident that the costs will decrease very rapidly.
Karrberg: Yes, the systems will be expensive at the start and come down in cost in the following years, but you should also know that you save costs on fender benders, car insurance is likely to go down, and also fuel economy will be improved.
Gus Bilirakis (R-Fla.): Where does the work on V2V communication fit into the overall blueprint of deploying self-driving cars?
Pratt: Vehicle to vehicle as well as vehicle to infrastructure communication is of critical importance to autonomous vehicles. Of course, we drive using our own eyes to see other vehicles, but the potential is there for autonomous vehicles to use not only the sensors on the vehicle itself, but also sensors on neighboring vehicles in order to see the world better. And so, for example if you’re going around a corner, and there’s some trees or a building that’s blocking the view, vehicle to vehicle communication can give you the equivalent of x-ray vision, because you’re seeing not only your view, but also the view from other cars as well.
It’s going to be pretty hard to make a vehicle that’s safe in all conditions. That’s this Level 5 vehicle that we keep talking about. And the standards may be very high, because again, it’s a machine, it’s not a human being, so our ability to empathize and forgive will be low. We have to give ourselves every possible tool in the tool chest in order to try and solve this problem. So I think that vehicle to vehicle and vehicle to infrastructure is extremely important, and that saving spectrum for that use is also very important.
Bilirakis: In Lyft’s view, what are some societal and economic benefits that we could expect to see from the deployment of self-driving cars?
Joseph Okpaku, Lyft: We often talk about the benefit that Lyft in its current form as a ridesharing platform has financially for drivers, but one of the things that I think often gets lost in the conversation is how important transportation is for economic upward mobility on the passenger side, meaning that one of the biggest factors for economic opportunity is access to reliable and quick transportation.
We’ve already seen some of the impacts we’ve had, we believe, on the customer side, just by providing safe and quick and reliable options to jobs, to get to and from work, that previously didn’t exist. So, if you buy that concept, and you apply it across a grand scale that an AV platform can provide, then I think the economic opportunity that it confers is really significant, and it can really help a lot of people who are in economic need get to and from their jobs [without having to] rely on insufficient public transportation options. In addition, [there is] the ability for non-emergency transportation. We’ve seen ridesharing start to partner with organizations on that front already; I think the ability to do that on a greater rate, or more efficient rate, expands once you include autonomous vehicle technology into the mix.
Evan Ackerman is a senior editor at IEEE Spectrum. Since 2007, he has written over 6,000 articles on robotics and technology. He has a degree in Martian geology and is excellent at playing bagpipes.