The self-driving car has captured our imagination.

Experimentation with the technology has been around for decades, but recent advances and particularly the advent of the Google Self-Driving Car in the past few years have pushed the technology to the front of public awareness. In 2011, Nevada was the first state to pass a law permitting autonomous cars on roadways. Florida, California and Michigan have since followed suit. The United Kingdom recently did the same.

In the news media, Atlantic Media’s CityLab has devoted many stories in the past year to what a driverless world would mean, touching on everything from government savings to insurance liability.

Locally, former Madison Mayor Dave Cieslewicz opined that “driverless cars will change everything” (not for the better if you’re a cab driver or a UPS guy), saying the city needs to start planning for a future that doesn’t include human drivers. He asks why we should spend public money on 1,200 parking stalls downtown for the Judge Doyle Square project when eventually no one will need to park cars away from home.

Perhaps all of that will happen someday, but visions of being able to step into a driverless car outside your front door to get to work while spending the whole time on your phone are still a ways off, says Teresa Adams, a UW-Madison professor of civil and environmental engineering.

She recently finished a three-year stint on a U.S. Department of Transportation committee that advises the secretary of transportation on “intelligent transportation systems,” a broad field of inquiry that includes driverless cars.

In a recent interview, Adams explained that limitations of the radar and lidar (light-sensing radar) systems on these cars will make navigating in high-congestion and less-regulated areas (think crowded parking lots) a high bar for autonomous cars to clear — at least for the near future.

But that’s not to say that big changes aren’t on the way soon behind the wheel. In particular, Adams said car-to-car communication safety features are probably going to arrive in less than a decade and will likely have a major impact.

“The bottom line is that we’re at a real turning point in how we’re going to relate to our vehicles and how our vehicles are going to provide for us,” she said. “We just have the technology now.”

The Capital Times: So, are driverless cars coming to our driveways, and if so, how soon?

Teresa Adams: I think that technology is coming and will likely be in our vehicles. It’s hard to say when, because the cost of the technology is still really very high. I think that when we do see it, it’s going to be more on highway routes, where engaging in the traffic is a little more controlled.

So, in places where there are fewer variables?

Exactly. If you’re doing a lot of weaving or you’re trying to get around an incident site or something like that, the cars just aren’t smart enough yet. They’re just not going to replace the human — not yet anyway. So when we talk about when these cars will be in Madison, if you’re thinking about in our communities, it’s not going to happen for a while. But if you’re thinking, say, around the Beltline, it could be.

So you see this being used most likely on roads dedicated to driverless cars?

Or lanes that are dedicated to it, so that all the cars are behaving nicely, playing nicely.

Generally speaking, how does the technology work?

These cars have radar and lidar sensing what’s around you and how close it is and what the terrain is. So it’s the vehicle sensing its surroundings and then, of course, it has a path. Somebody’s going to have to tell it where to go, just like GPS. And then from there, it has to know properties of the highway, turning restrictions, speed limits and things like that that are on databases.

Would there be special infrastructural needs for driverless cars?

If we ever want to implement something like this, we have to be careful and make sure that the striping on the roads is really in good shape because these things are going to depend on that. It will be sensing to stay in its lane, so it’s got to know where its lane is.

Would this be harder to do then, in cities like Madison that often have snow on the roads?

That could very well be a problem. Another potential problem is where the sensors are mounted. In the past some of them have been mounted on the outside of the vehicles, but we know they can get gunked up. So now we put them on the inside of the vehicles where it’s protected a little bit more, but it’s still not perfect. So, that’s a good point. All the demonstrations I’ve seen have been in good weather.

In the nearer future then, what kinds of technological changes do you see on the horizon?

The other model is the connected vehicle model, where the cars are communicating with other cars. It’s usually still the driver doing the driving, it’s just that the connected vehicle part is providing a lot of enhanced safety. So, telling you when there’s an imminent collision, if you’re going too fast and you’re going to slam into somebody in front of you. We have a lot of technology similar to that — adaptive cruise control — right now, but that’s not being done by communicating with a vehicle, that’s our own vehicle sensing what’s in front of it. This would be where the cars would talk to other cars.

Do you think the government is likely to require this technology in cars at some point?

I think the safety benefits are being proven — the number of crashes that could be prevented if we had these devices where the communications are a lot faster than a human can think. I actually got a chance to sit and drive in one of these vehicles and experience some of the interfaces. As soon as you put on your blinker with your intention to change a lane, another vehicle is communicating, ‘Hey, I’m here.’ The difference is then in how the auto manufacturers will have the cars communicate that information to the driver.

So cars will be talking to drivers?

I’ve seen ones where a row of lights flashes across the windshield, which would tell you you’d better stop. I’ve seen ones where, to prevent a lane change collision, the driver’s seat would vibrate on one side. I haven’t seen ones that talk to you. It’s quicker, more intuitive.

How soon will we see this technology, do you think?

I have talked to people who say it could be within five to seven years. Even the next car you buy is going to be really different, and it’s just going to keep going.

Managing Editor of The Capital Times

You might also like

(2) comments


Automated systems have two much larger hurdles to overcome. 1) They fail to include communication between drivers. In other words, when you're driving you often 'read' what another driver intends to do, or what they want, according to how that person is driving their car. Also you read nonverbal communication - will the computer?. Another driver waves at you to 'go ahead' at a 4-way stop. Do you trust that person? Or, why would you trust the automated system to decide for you? 2) And how will the computer/driverless car act according to your ethics? Eg. If a child runs out into the road do you swerve into the opposite lane where you will be crushed head-on by a semi truck or do you run over the child? Whose at fault in either case? Who pays what?


with all the recalls coming at a fast pace this year - I would like a car with minimum electronics - hell - I would even go for crank windows. one less thing to go wrong and cost money

Welcome to the discussion.

Keep it clean. Exchange ideas and opinions on posted articles. Don't promote products or services, impersonate other site users, register multiple accounts, threaten or harass others, post vulgar, abusive, obscene or sexually oriented language. Don't post content that defames or degrades anyone. Don't repost copyrighted material; link to it. In other words, stick to the topic and play nice. Report abuses by clicking the button. Users who break the rules will be banned from commenting. We no longer issue warnings. Use the 'Report' link on each comment to let us know of abusive posts.