BDTI: What about passenger vehicles? Do today’s vehicles offer any automated driving features?

Prof. Varaiya: Some high-end vehicles offer adaptive cruise control and lane keeping. Adaptive cruise control is meant for long-distance country driving, and it assumes you maintain a large gap between cars. There is a lot of interest in developing signal processing algorithms that will allow use of adaptive cruise control in stop-and-go traffic. In this case you are talking about much smaller distances between cars, so you need much more confidence in the system.

Adaptive cruise control provides automated control—it actually speeds you up or slows you down. Lane keeping systems don’t currently provide any control; they just give an audible warning to the driver or increase the steering resistance. This is because the liability issues associated with automated steering are enormous.

BDTI: When will we see fully automated vehicles on the highways?

Prof. Varaiya: Autonomous steering will most likely start in trucks, for a number of reasons. First, many U.S. states are considering implementing truck-only lanes. It is much easier to deploy automation when you only have one type of traffic to deal with. Second, it is much easier to justify the cost of automation when the vehicle itself is expensive, which is the case for large trucks. Third, you may be able to eliminate truck drivers, and drivers are very expensive. So the trucking industry—particularly in the U.S. and Europe—has a lot of interest in automation that goes beyond cruise control and lane keeping.

BDTI: Besides exclusive lanes, what is needed to make autonomous highways possible?

Prof. Varaiya: The main challenge is building a system that prevents unauthorized vehicles from entering the exclusive lanes. You also have to figure out how to deal with accidents. If the autonomous lanes are not separated by a physical barrier, then accidents in the non-autonomous lanes may spill over into the autonomous lanes. You also have to figure out how to deal with accidents or breakdowns in the autonomous lane itself.

BDTI: Is autonomous vehicle technology mature enough for widespread deployment? Are economics the main constraint?

Prof. Varaiya: In terms of the purely technological aspects, I think that many of the problems are solved. However, there are no data on reliability. There is nothing to show that the equipment itself is reliable, let alone that the systems are reliable enough for a large-scale deployment.

On the economic question, it is difficult to imagine mixing autonomous vehicles with manually driven vehicles because the behavior of manually driven vehicles is unpredictable. Plus, most automated systems assume vehicles can communicate with each other, but manually driven vehicles are not going to be in communication.

This means you need exclusive lanes for the automated vehicles. But to justify the cost of exclusive lanes, you need a large population of vehicles that can use these lanes. This is a chicken-and-egg problem, and I don't know how it will be solved.

BDTI: What else needs to happen before we have fully automated highways?

Prof. Variaya: One issue that is always raised when this topic is discussed is liability. Who is liable if there is a collision? The automaker? The automation system supplier? The driver? It's similar to the problem we had with fly-by-wire controls on aircraft. In the U.S., Congress had to pass a liability limitation law before anybody would pursue that technology.

BDTI: How will automated highways affect people’s daily lives?

Prof. Varaiya: Most likely, this technology will be introduced through truck and bus lanes. This will make the passage of trucks and buses much easier. It will also separate heavy vehicles from passenger vehicle traffic, which will improve safety and road capacity. And of course there will be savings in pollution and gas consumption.

When the technology is introduced in passenger vehicles, the impact will be dramatic. It will be a whole new mode of transportation, something like having a personalized train. We believe collisions and fatalities will be reduced. And we might finally see the end of road rage!