Self-driving cars are a lot closer to reality than most people think. Auto makers are promising cars that won't need a human driver behind the wheel within the next four years. But a whole lot needs to happen between now and then. We caught up with California Department of Transportation Director Malcolm Dougherty during this week's Automated Vehicles Symposium in San Francisco to talk about how the state is preparing.
Take Two: What will California need to change about its infrastructure to accommodate self-driving cars?
Dougherty: There’s 36 different companies that have permits to test autonomous vehicles on public roads today, so those companies are learning as they’re going, but some of the issues that are coming up is that those vehicles are looking at the environment that they’re driving in — the condition of the infrastructure, the condition of our signs, the condition of the delineation is a factor in how effectively they operate within that environment.
How much money does the state need to make the roads ready for them to operate?
We just passed SB1, the new transportation revenue package. In my eyes, that’s a sufficient amount of resources for cities, counties and the state to properly maintain our existing infrastructure whether it’s roads, bridges, drainage systems like culverts, but it also identified intelligent transportation systems as one of the important items we need to maintain. That would be signals, cameras, changeable message signs, ramp meters and those things, so I think that there is adequate funding.
What about for vehicle-to-infrastructure technology?
As you start to put out components in the field that are talking to cars or signals that are sending out information for cars to receive, that’s a whole other contemplation for us. All of our signals already have technological cabinets that are controlling the timing and those types of things, but it would add a little bit of a cost to add that to all of our signals. I may not need to put up changeable message signs in the future. I may be putting up boxes that are sending out a signal to the car that tells the automobile what the downstream information is as opposed to me trying to relay it to them in three lines of text.
Do you consider yourself an advocate for self-driving cars?
I am and for a couple of very important reasons. I think the safety implications are significant. If my car is talking to your car, you and I are much less likely to run into each other when we approach an intersection. If any of our cars are looking out for pedestrians and bicyclists, I think that will be a very big safety benefit from the vulnerable user on the transportation system. I think if my car is talking to the signal, I’m also less likely to run a red light by accident. We’re having a lot of accidents with human-driven cars and we’re having a lot of fatalities, so I think the opportunity for the technology to improve safety is significant.
How long has California been testing self-driving cars?
The first real field test we did was 20 years ago in 1997 on Interstate 15 in San Diego on HOV lanes. Those cars were using human drivers, and steering, but they were using radars to pace off the vehicles in front of them and also following magnets in the pavement. Back then, that was very innovative. Today, we know we can’t put magnets in the pavement everywhere for cars to follow so now the modern car that has equipment like this is following the lane lines, using cameras, so that puts the onus back on us to make sure that delineation is very clear and discernible.
How do you delineate clearly?
One area this comes into with California is we’ve been using those Botts' Dots. The automated vehicles [AVs] can follow lane lines. They can’t follow the Botts' Dots, so we’re actually changing our delineation standards to go away from the Botts' Dots which we’ve been using for decades because AVs have a difficult time following those.
What other parts of the road need to be changed?
All of our lane lines are going to get thicker. Today our lane lines are only four inches thick. Now every lane line we lie down going forward is going to be six inches thick. I’ve already started to see some of this transition. That’s good for the AVs, but it’s also good because it will be much more clear delineation for the human driver as well.
How many miles of road need to be changed?
We have 50,000 lane miles in California. We’ll do it systematically as we have construction projects and we are constantly re-striping our highways all the time anyway, so it will take us a couple years to transition over. But that material only lasted a few years anyway, so we were constantly refreshing it. I would say we would at least prioritize the interstates and freeways and have that done in the next two or three years.
What about HOV lanes?
I could see a conversation getting struck up at a certain point in the future where we’re not just talking about HOV lanes. We’re talking about autonomous car lanes. I anticipate it’s coming because I only have so many lanes out there. There's a limited opportunity to build new lanes. That may not be the most strategic way to add capacity to the transportation system anyway. But it would be a very controversial question to take one of our existing lanes and make it an AV lane. So that, I don’t know if it’s going to happen, but I’m anticipating that the conversation is likely to come up. Folks will talk about AVs maximizing throughput. You’ll be able to put more cars faster and closer together when they’re autonomous. Well, they all have to be autonomous to have that contemplation. They have to be connected. So you can’t have a human car in that scenario. You would have to have separate lanes to be able to drive multiple vehicles, close together, at fast speeds, which we’ve tested with truck platooning with the same technology.
How would that impact EV access to the HOV lanes?
There’s an anticipation that a lot of the AVs will end up being EVs also. I’m worried about through put. I can’t give up the HOV lanes and put in autonomous single-occupant-vehicle lanes. I’ve lost volume, so it may be that there’s a combination of both. Maybe it’s autonomous HOV lanes. It’s got to be something, but I’ve also got to think about congestion, and I’ve got to think about through put, but I can see that conversation coming up.
How will self-driving cars affect traffic?
The interesting thing and an outstanding question on AVs is, "Is it going to increase or decrease vehicle miles traveled?" and that’s an open question. We don’t know the answer. But as an owner/operator of a transportation system, I have a concern about it increasing Vehicle Miles Traveled because I don’t have the capacity to handle that. We already don’t have enough capacity for the number of miles that are being driven today. If people are driving more because it’s autonomous or if you have empty cars taking dead load trips to go back and forth, you may actually be increasing the amount of VMT. You also have shared mobility that may not have a driver in the future, so at the onset of AVs and shared mobility, is VMT going to go through the roof? That doesn’t meet some of our other objectives and it actually may cause problems, but it does provide access and it provides mobility, so it’s going to be a balance. There’s a lot of outstanding policy questions.