Viewing the Road Ahead for Self-Driving Cars
This week, General Motors announced that it would pour $500 million into the ride-sharing service Lyft, with an aim of eventually producing a fleet of self-driving cars. And the Consumer Electronics Show in Las Vegas was filled with autonomous vehicle tech tidbits from companies such as Toyota and Nvidia. But what might a future in which all cars can drive themselves do to our cities, towns, and society? Industry observers say that while it’s clear that there will be robotic cars, it’s much less clear how people will choose to use them.
We wanted to know what you thought, so earlier this week we sent out a survey to our social media fans. We asked them to consider the following scenario:
“Imagine a world where every car is autonomous, and no one drives. You can either own your own car, or share from a pool of vehicles operated by a service like Uber or Lyft.
Being part of a sharing service would be cheaper, but the car might take longer to get to you than one you own yourself.
In this case, would you rather own a car, or share it?”
Over 550 people responded – and the results are in!
Most of our fans would rather own an autonomous car than share one as part of a service.
We also asked for zip codes, to see if there were any patterns among the respondents.
Do you see any patterns? What do you think? Let us know in the comments!
Chris Ziegler is Deputy Editor of The Verge in New York, New York.
David King is an assistant professor of Urban Planning in the School of Geographical Sciences and Urban Planning at Arizona State University in Tempe, Arizona.
Susan Handy is Director of the National Center for Sustainable Transportation at the University of California – Davis in Davis, California.
Bryant Walker Smith is an Assistant Professor in the School of Law at the University of South Carolina, and an Affiliate Scholar in the Center for Internet and Society at Stanford Law School. He’s based in Columbia, South Carolina.
IRA FLATOW: This is Science Friday. I’m Ira Flatow. If you look around the exhibit floor at the Consumer Electronics Show in Las Vegas this week, you might have thought that you took a wrong turn and ended up at the Detroit Auto Show. Transportation systems were big, from tiny electric scooters to plans for entire fleets of autonomous cars.
Chris Ziegler is deputy editor of The Verge. And he’s been at the CES show all week. Welcome to the program. Boy, you must be tired, Chris.
CHRIS ZIEGLER: Oh, you’re not kidding. It’s been a week, and I’m– like you said, we go right into Detroit.
IRA FLATOW: That’s right, the next week after. So tell us about the big trends you’re seeing in. The autonomous cars are making a big splash, so to speak?
CHRIS ZIEGLER: Autonomous cars are big here, as they have been basically every CES for the past three or four years. This is sort of seen as the place to show the tech side of the auto industry. Whereas, Detroit and New York and LA, some of those other big car shows, are really more focused on the design and horsepower– a very classic sort of car news opportunity. But here, it’s really about the tech.
IRA FLATOW: Yeah, and what about electric cars? Have they taken a backseat? Or are we seeing more of them being shown?
CHRIS ZIEGLER: Electric is big here too. Electrification, connected car, autonomous, those are the big three. And with electrification, specifically, of course, GM probably made the biggest news here with the introduction of the Bolt, which is their production high range, relatively affordable EV that’s going to be hitting the market later this year. And Volkswagen is trying to make good on its diesel problems with the BUDD-e, it’s called B-U-D-D dash e, which is a throwback to the classic Microbus but with an electric drivetrain.
IRA FLATOW: Cars are the next big space for the internet of things, right, Tethering all of our devices together. Are you seeing where cars are going to be interconnected, talking to each other on the roads?
CHRIS ZIEGLER: Yeah, so there’s a technology called vehicle-to-vehicle, V2V for short, that will be debuting on production cars this year, where the cars on the road can sort of make an ad hoc mesh network and talk to each other and what each other know when, say, one car’s braking really hard but you can’t see it. That notification can come into your car, so that you’re more prepared to break hard, if there’s an emergency ahead of you, for instance.
But then going back to the internet of things more generally, there are a couple interesting announcements here. Ford is playing around with working on the Amazon Echo. So for instance, when you’re sitting in your house, you can say, Alexa, what’s the battery level in my car? And then your Echo, the little voice box in your house, will tell you whether your car is charged or not.
IRA FLATOW: Well, you have to have a car that’s charging in the driveway or garage or something.
CHRIS ZIEGLER: That’s right.
IRA FLATOW: I like that. Has there been any talk about the GM-Lyft partnership that was announced?
CHRIS ZIEGLER: Yeah, that’s been a surprisingly low key announcement. I thought that was going to make a big splash here and people we’re going to be talking about it a lot. But I think that, cynically, it’s seen as a very defensive maneuver that doesn’t necessarily have any short term technology implications, because Lyft doesn’t bring a lot of self-driving experience to the table, not as much as Uber, which has its own self-driving R&D lab in Pittsburgh.
And in the very short term, this might be more of an opportunity for GM to just sell vehicles into Lyft’s fleet. But we’ll see. We’ll see if it actually turns into something bigger long term.
IRA FLATOW: Well, we hear autonomous cars– they’ve been a feature of the CES for a couple of years now. But are we seeing any– something radical that is going to move– that actually move the dial a bit?
CHRIS ZIEGLER: No, and that’s my concern. The term that I’ve been using here is that we’re in a hype gap, where there is a protracted phase of hype around autonomous cars and not a lot of substance, not a lot of things that people can really sink their teeth into. They can’t take a self-driving down the road. But it’s trickling out.
We have things like Tesla’s Autopilot, which is now in beta in cars that are on the road. GM will be debuting something this year, called Super Cruise, which is a similar feature, where when you’re on the highway, you can press a button and let the car drive itself. But in terms of announcements here, they’ve been pretty low key.
Ford announced that they’re using a new higher resolution sensor, for instance, in their research self-driving cars– so very incremental stuff. And then I think, toward 2020 is one we’re going to see this huge moment of a bunch of automakers releasing a bunch of at least partially autonomous vehicles onto the road.
IRA FLATOW: Yeah, well, we’ll be talking about it a little bit more. But I want to wish you stay healthy, walk those floors, and good luck at the Detroit show next week, right?
CHRIS ZIEGLER: Thank you very much. Yes, it’s going to be a good time.
IRA FLATOW: Yeah, we’ll see more of that stuff there. Chris Ziegler is the deputy editor of The Verge. And he has been at the CES show all week. But if self-driving cars do become a widespread thing– think about that– how is that going to change the way you get around, or the shape of our cities, or our public transportation.
Joining me now are two people who think a lot about transportation and how it fits into the fabric of our lives. David King is an assistant professor of Urban Planning at Columbia University here in New York. He’s here in our New York studios. Welcome to the program.
DAVID KING: Thank you for having me.
IRA FLATOW: Susan Handy is director of the National Center for Sustainable Transportation. That’s at the University of California at Davis, as they say, UC Davis. Welcome to Science Friday.
SUSAN HANDY: Delighted to be here.
IRA FLATOW: All this talk about autonomous driving cars, David, how easy is it to imagine where all of this is going? Or are we just pie in the sky at this point?
DAVID KING: Right now, I think it’s mostly pie in the sky. We’re very good at predicting the direction of technology. And we can be very confident that we’re going to have self-driving cars in the not-too-distant future. We’re very bad at predicting how people are going to use them. And ultimately, until we know how much they cost, how much a trip costs, how we price our roads, how we adjust our parking, and so on, we’re really not going to have a good idea as far as how autonomous cars affect our cities.
IRA FLATOW: So now would be a good time to think of the macro picture, right?
DAVID KING: Exactly. Now is a good time to think of the macro picture. And that doesn’t mean that we have to regulate these cars. But we need to think about what’s the role of our streets, what do we want from our streets.
One of the things that we’re trying to do now in cities is we’re trying to slow down traffic. And we’re trying to create streets where people can play and they could walk and they can bike, they could be comfortable. Those are not necessarily streets that we’re going to want any robot car to decide is a good street to drive down. And that’s one of the issues that’s happening.
We’ve seen this with Waze in Southern California. Waze is a traffic route app.
IRA FLATOW: Yeah, I use that all the time.
DAVID KING: And they will route people off the freeway through residential neighborhoods. And these residential neighborhoods have quiet streets, and they expect quiet streets. But the algorithm doesn’t care. A road is a road is a road. And there’s a lot that we have to rethink, as far as how we classify our streets and mostly how we charge for access to our streets.
IRA FLATOW: Susan Handy, is the autonomous car going to replace our cars? Or is it just going to be a separate additional mode of transport?
SUSAN HANDY: Well, I think as David said, it’s really hard to predict exactly how this is going to play out in terms of how people will respond. I think it’s probably a mistake to think that this is going to revolutionize everything. Likely, we’re going to see some evolution over time as to how autonomous vehicles fit into our transportation systems.
And it could go some very different sorts of ways. It could end up being kind of an individual ownership model, where every household has its own autonomous vehicle. Or it could be much more of a shared model, where we have privately or publicly owned fleets of autonomous vehicles that give individual rides or even shared rides. And how it involves has huge implications for our cities and for the environment.
IRA FLATOW: Especially since people choose where they live by sort of a commuting distance to a central hub. If you have an autonomous car that sort of mimics a train, because you get in, you can do your homework, you can– won’t that create more suburban sprawl, Susan, moving outwards from where you are normally spending your day?
SUSAN HANDY: Yeah, well again, I think we need to be careful in thinking that this necessarily leads to one kind of future or another. But it certainly does make it easier for people to live farther from where they are working. And so to the degree that there are people out there who would like to do that, then autonomous vehicles could help to facilitate that.
On the other hand, we are seeing evidence that there’s an increased interested in living in denser urban environments, the kind of streets that David was talking about. And in that case, autonomous vehicles may play a very different role in the lives of people living in those kinds of environments.
IRA FLATOW: David, let’s– go ahead.
DAVID KING: Well, I was just going to say, again, until we know how we charge for these, if you’re paying by the mile or paying by the minute, you’re still not going to live way out in the exurbs and then use this to commute into the city center. And regardless of how the technology works out, an hour spent in a car is going to be less enjoyable than an hour spent at home or at the movies or out to dinner or something like that. So we’re going to see people change their behaviors based on how the technology evolves.
IRA FLATOW: But your car that’s in New York, for example, and you know what traffic in New York is like, the pedestrians in New York play chicken with the traffic.
DAVID KING: Mm-hm.
IRA FLATOW: They do not pay attention to the rules. They’re terrific jaywalkers. Can you design a car that’s not going to play chicken with the pedestrians? Or the car will never move, right? But it always sees pedestrians crossing the street.
DAVID KING: So I think it’s plausible, maybe not probable, but it’s plausible that autonomous cars will require even more road rules to our roads, that pedestrians and cyclists and trucks and all these other modes are going to have to be very constrained in their movements so that the robot cars and the robot trucks and everything else are able to move freely. That’s one possible scenario. The technology is such and developing so quickly, I don’t think we can say with any certainty how that will work out.
IRA FLATOW: And you know, I was thinking of, there’s a traffic cop. Does the car know that there’s a traffic cop directing traffic and the lights are not working anymore? I would imagine you would have little transmitters– we were just talking here, heard about the show, the car show, where the cars are talking to each other, not only would the cars be talking to each other, but it might be talking to a FedEx truck that’s sitting there, knowing that it’s figuring out there’s a chip in that telling it it’s sitting there.
DAVID KING: Yeah, these are all plausible scenarios. And right now, there’s sort of an idea that there’s going to be a benevolent supplier of the cars. And then there’s going to be all these other technologies that go on. And we really don’t have any idea how they’re going to work together. But it’s very much the case that the future of our roads is in flux right now.
IRA FLATOW: Susan, you know, there are a lot of old versions of the future. You go to these future works or World’s Fairs and things, and they have streams of identical car pods moving along, sort of attached to one another, like a train. Does that seem likely?
SUSAN HANDY: Yeah, I think the possibility that autonomous vehicles could turn into a public transit system like we’ve seen in some of those older visions is pretty interesting to think about. Whether it’s going to happen that way or not depends on a lot of decisions we make about how we regulate them and what sort of public investments we make in this innovation. Whether it’s likely, I don’t know.
IRA FLATOW: Our number, 844-724-8255. This is Science Friday from PRI, Public Radio International. I’m Ira Flatow. Let’s go to the phones, to Bill in Pensacola, Florida. Hi, Bill. Bill, are you there?
IRA FLATOW: Hi, go ahead.
BILL: Hey, this autonomous vehicle discussion over the past couple years seems to completely ignore the opinions or interests of people like myself, who are automotive enthusiasts, people who love cars, work on cars, restore classic cars, want to drive them and experience the driving process. And what’s particularly frightening is what is the– when driverless cars reach some sort of critical mass, they’re obviously going to have some effect on the roadway system. And it may be such that people like me can’t go out for a Sunday drive and enjoy the process of driving, because the roads are designed almost like railroad tracks than open highway.
IRA FLATOW: Good point. David.
DAVID KING: Well, it was a few years ago that James May of the Top Gear show said, in the future, everybody’s going to drive Ferraris. And what he meant by that was, when autonomous cars come around, the only people who drive are going to be enthusiasts like yourself. And they’re going to drive cars that are really fun to drive. And that becomes an issue.
And again, how that works out in the future depends on how we regulate our roads and our streets and how we pay for them, that if these are a source of commerce and if there’s businesses that depend on having our traffic flow freely, then it might not be that you’re able to drive in cities doing that. But in rural areas, I suspect, there will always be open roads.
IRA FLATOW: Susan, what factors go into people’s transportation choices? How personal are these decisions?
SUSAN HANDY: Well, one theory of travel behavior says that people maximize their utility, that people choose the option that makes the most sense to them. And of course, that’s going to differ for everyone. So some people put most weight on the monetary cost or the travel time. But we know that there’s also this element of enjoyment.
So that’s going to factor in as well. And different people are going to make different choices, depending on what their own needs and constraints and preferences are. So that’s partly why it’s so hard to predict how this is all going to play out.
IRA FLATOW: Yeah, we have a whole generation of kids– and I mean people, kids 5, 10 years old, who have grown up in the backseat of a car, just playing video games or a working on their cell phones or something like that– who don’t seem to care at all about the drive. Whereas, when we were kids, we want to be in that front seat, doing to drive. Is this going to help just move that whole population to self-driving cars, because, look, I’ll be able to do what I did as a kid, and I want have to change my habits.
DAVID KING: I think a big issue there is, do they want to drive or do they just want the access and the mobility? And if they just want the access and the mobility, we can continue to offer that through self-driving cars of some sort. And they don’t ever have to learn to drive. And that would be wonderful in terms of safety on the roads, that if we can have robots driving our cars, our roads will be safer.
And I think that that’s a real issue, that kids probably want to maintain that mobility and that access. And they may not care about the actual physical act of driving. But we’ve been making driving easier over the decades through cruise control and automatic transmissions and self-parking now and all these other things anyway.
IRA FLATOW: But you know, it’s going to take a while, no matter what happens. I had a discussion with the executive at Toyota a few months ago. And I said, the biggest selling car in America is the Ford F-150 truck. I don’t think those people are going for autonomous driving any time soon.
DAVID KING: Well, to relate this to the question that you asked Susan earlier, there’s going to be a lot of market segmentation in this, that a pickup truck is not going to be easily replaced by an autonomous taxi. And there’s just no way that that’s going to happen. And you’re still going to need these many different types of vehicles.
IRA FLATOW: All right, we got to take a break. When we come back, we’re going to talk lots more about self-driving cars. And we’re going to expand the discussion to talk about who’s legally responsible in a fender bender between two self-driving cars. Can you sue the computer? I don’t know. That’s something that the– I’m sure the insurance companies are thinking all about that now.
Stay with us. Our number, 844-724-8255. We’ll be right back after this break. Stay with us.
This is Science Friday. I’m Ira Flatow. We’re talking this hour about autonomous vehicles, the coming age of self-driving cars. My guests are David King of Columbia University and Susan Handy of UC Davis.
And when it comes to self-driving cars, one of the big unknowns centers around the law and liability. So when a car that’s driving itself gets into an accident, who’s at fault. If two autonomous vehicles get into a fender bender, who pays– the computer driving the car, your insurance company?
Bryant Walker Smith is an assistant professor in the School of Law at the University of South Carolina and an affiliate scholar in the Center for Internet and Society at Stanford Law School. He’s based in Columbia, South Carolina. Welcome to the program.
BRYANT WALKER SMITH: Thank you.
IRA FLATOW: So the big question on most people’s minds is probably one of liability. Who’s liable? Or have we not figure this out yet?
BRYANT WALKER SMITH: And it puzzles me sometimes why that is one of the most interesting questions, because we know who suffers when there’s a crash. It’s the person who’s injured. And then the question is, who ultimately bears the cost for those injuries? And it’s everybody.
We do through the purchase price of the self-driving cars, or the driverless car ride we might take, or the purchase price of the package that might be delivered by a driverless vehicle, or the price we pay for our insurance or our medical care. So in between the question of how we get from crash victim to ultimately us all paying is an important question. But it’s one that might concern the companies more than individual consumers.
IRA FLATOW: You mean, the car companies?
BRYANT WALKER SMITH: Yes.
IRA FLATOW: Them being–
BRYANT WALKER SMITH: The companies that are developing these technologies.
IRA FLATOW: And it sort of reminds of a question I was talking about the other day is, if we get autonomous– if we have robots, flying robots delivering packages and it drops on your head, the package, instead of on your lawn, isn’t that sort of the same kind of question we’re asking?
BRYANT WALKER SMITH: It is. And I will tell you, if you have to be hit by a car, and no one wants to be, you are better off getting hit by an automated vehicle. And the reason why is because the companies and developers and users of those vehicles are at least initially likely to have deeper pocket than your ordinary negligent driver.
IRA FLATOW: The ones that are driving out, the test cars that are out there now, who’s liable for those?
BRYANT WALKER SMITH: Hm. So if one of those were to be at fault in a crash, the people injured in the crash could turn to the individual drivers, but would much more likely turn to the developer of the systems and the company actually conducting the testing, whether it’s an automaker or Google or a university, and say, hey, you put out a system on the road that injured me through its failure. Pay up.
IRA FLATOW: Well, let’s talk about the car itself and the ethics of some of the things the car might decide to do. Let me give you a hypothetical question. Let’s say, in a case where an accident seems imminent and people have to make quick choices where to swerve– a dog runs out into the middle of the street, let’s say, and you either have a choice, if you’re driving, of hitting the dog or hitting another car next to you, or maybe some other– or hitting a tree, who decides what the car chooses there? And how does the car make that decision?
BRYANT WALKER SMITH: Well the first question is, was the vehicle going too fast in the first place? Had it been going at a slower speed, could it avoid that so-called dilemma situation altogether? Drawing the boundaries of our analysis larger is one of the things that automated vehicles may well do, so that we get into these situations less frequently.
But let’s assume, hypothetically, that an automated vehicle were to encounter this choice of bad situations. Well, depending on the particular programming and the particular level of automation we’re talking about, it may simply try to take as much energy out of the system as possible– brake, brake, brake really fast. That may avoid the dog. It would certainly avoid the other car. It may not avoid the dog. But it might be justifiable legally and ethically.
If it’s a more sophisticated car, then it’s going to have more decisions to make. But those will always be checked by uncertainty. We won’t really know, and a car won’t really know the result of all its decisions until it’s taken them.
IRA FLATOW: Well, what if I give– I am given the option for that car to choose. Like, you put options into your computer, what if I say, if it comes down between hitting a dog and hitting the car– and I want to put this crassley– I’d rather hit the dog, because I could kill somebody hitting the car. Can I make that as an option in my software that runs the computer, let’s say?
BRYANT WALKER SMITH: Mm-hmm. Well, now you can’t.
IRA FLATOW: [LAUGHTER].
BRYANT WALKER SMITH: In some systems that exist in production today, you can adjust, perhaps, the aggressiveness of a system. So you’re following a vehicle closer than you might otherwise. You’re following faster than you might otherwise. Those are options.
As the systems get more sophisticated, might it be possible that more user preferences could be integrated? Yes, but subject to some limits. I don’t think you’ll ever get to say, I want to just hit every dog I can find.
IRA FLATOW: Bringing in my other guests, David King, Susan Handy, David, what do you think about these choices?
DAVID KING: Well, I think on this last point, this is a really big issue as to who owns the code for these cars, who owns the programming. And there’s recently some court cases, where the owner of the car, rather than the manufacturer, actually has the rights to make changes of the code on existing cars. And if that holds for autonomous cars, you may end up that they start driving in certain ways that reflect the personalities of the individuals. And that would be a very, very different type of– a very much more competitive market on the streets, which would be problematic.
IRA FLATOW: Susan, any comments on this?
SUSAN HANDY: Yeah, I think this is outside my area of expertise.
IRA FLATOW: [LAUGHTER]. Yeah, and then you’ll have the problem of hackers. People can hack into the code. They want to do something crazy to upset it. And then who’s responsible for that sort of thing? Bryant, any thinking about that you want to–
BRYANT WALKER SMITH: Yeah, so a intrusion into a system is another kind of failure. And if it’s one that the developer should have anticipated, could have anticipated, could’ve prevented, then it’s a pretty clear case of liability. In some cases, even if no physical damage occurs, companies could be liable for the diminished value of their products or for representing the safety or security of their systems is better than it actually was.
IRA FLATOW: Let me go to the phones the few minutes we have left. Let’s go to Bruce in Mountain View, California. Hi, Bruce.
BRUCE: Hi, Ira. Good morning. It’s such a pleasure to be on your show. I actually cannot believe this. Listen, I am a driver for Uber. I’ve been doing it for about a year and a quarter now. And I have a lot of doubts about driverless cars, just by knowing the people I pick up.
People are babies. They change your destinations all the time. They want you to stop here and there. There are so many factors involved with this technology. I don’t think it’s going to be ready.
It may be ready, and there’s a lot of hype like you’re talking about. There’s a big hype gap there. But I don’t think people as a public are going to be ready for this for a while. Now, I may be wrong, because I pick up a lot of Stanford students and none of them have cars anymore.
And as you were saying, back in our generation, that’s all we wanted when we were teenagers. We wanted to get behind the wheel. We wanted cars. We couldn’t get our license quick enough.
Today, I don’t get that feeling. People– kids don’t want to drive cars. And they’re back there and they’re content on having the flexibility to go out and drink and do things. But my gut feeling is, there’s just so many problems with this car– with these cars coming out in terms of the responsibility, for instance, if it gets hit. The amount of pedestrians– I mean, driving in San Francisco with just the number pedestrians, I just cannot see.
IRA FLATOW: All right.
BRUCE: –the driverless cars doing it.
IRA FLATOW: I get it. I’m here in New York. I understand what you’re saying. Thanks for calling. I mean, it’s going to take a little longer.
DAVID KING: It’s going to take a little longer. I mean, there’s a lot of cars on the road. There’s a lot of infrastructure and the built environment. And our cities are built around driving right now, for good or bad. And it’s going to take a long time to change that.
IRA FLATOW: Susan, do you think it’s going to take a generational change?
SUSAN HANDY: Well, I think there’s some generational change happening. But I think, as the caller points out, we’ve got a lot of important innovation going on right now. Autonomous vehicles, they’re somewhere down the road, almost surely.
But in the meantime, we’ve got a lot of other things happening that are very important that are expanding the transportation options that are available, whether it’s car sharing or ride sharing services like Uber and Lyft. We’re seeing innovation in bicycling with bike-sharing systems an electric-assist bicycles. We’re seeing new forms of transit service, some of which are privately provided.
And I think all of this, it’s expanding the options. And it is contributing to changes in how we think about transportation, that will especially be affecting the younger generations. So I think not even talking about autonomous vehicles, that we’ve got a lot of change going on that’s going to mean a very different transportation future.
IRA FLATOW: Final thought, David?
DAVID KING: Well, just to add to that, there’s also online shopping. There’s also delivery. There’s far more options to driving now than there ever have been. And it’s not at all clear as to, when we reduce driving, does that mean that people are going to continue to do the same amount of travel but in an autonomous car, or are they simply going to travel less and make fewer trips?
IRA FLATOW: Bryant, any last prediction at all about where we’re headed?
BRYANT WALKER SMITH: All those things that we think are going to be here tomorrow are going to take forever. And it’s the things we don’t even see coming that are going to surprise us.
IRA FLATOW: OK.
SUSAN HANDY: Exactly.
IRA FLATOW: Certainly, the philosophy of a New York Fifth Avenue pedestrian. David King, assistant professor of Urban Planning at Columbia University, Susan Handy, director of the National Center for Sustainable Transportation at the University of California at Davis, Bryant Walker Smith, assistant professor School of Law at University of South Carolina, thank you all for joining us today.
DAVID KING: Thank you very much.
IRA FLATOW: And happy driving to you all.