15.6 C
New York
Sunday, May 19, 2024

Teaching Self-Driving Cars to Watch for Unpredictable Humans

logoThe AI Database →


Human-computer interaction


End User





Machine learning

Machine vision

If you happen to live in one of the cities where companies are testing self-driving cars, you’ve probably noticed that your new robot overlords can be occasional nervous drivers. In Arizona, where SUVs operated by Waymo are sometimes ferrying passengers without anyone behind the steering wheel, drivers have complained about the robot cars’ too-timid left turns and slow merges on the highway. Data compiled by the state of California suggests that the most common self-driving fender benders are rear-end crashes, in part because human drivers don’t expect autonomous cars to follow road rules and come to complete, non-rolling stops at stop signs.

As for human drivers, some are nervous and scrupulous, others are definitely not. In fact, it’s even more complex: Some drivers are careful in some moments and hard-charging in others. Think: casual Sunday drive to the grocery store versus racing to get the kid before the day care late fees kick in. Robot cars might be smoother, and might make better decisions, if they knew exactly what sort of humans were driving near them.

Want the latest news on self-driving cars in your inbox? Sign up here!

Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory and Delft University’s Cognitive Robotics lab say they’ve figured out how to teach self-driving vehicles just that. In a recent paper published in the Proceedings of the National Academy of Sciences, they describe a technique that translates sociology and psychology into a mathematical formula that can be used to teach self-driving software how to tell the road ragers from the rule followers. Vehicles equipped with their technique can differentiate between the two in about two seconds, the researchers say, and can use the info to help decide how to proceed on the road. The technique improves self-driving vehicles’ predictions about human drivers’ decisions, and therefore the vehicles’ on-road performance, by 25 percent, as measured by a test involving merging in a computer simulation.

The idea, the researchers say, is not just to create a system that can differentiate “egoistic” drivers from “prosocial” drivers—that is, the selfish ones from generous ones. The scientists hope to make it easier for robots to adapt to human behavior, and not the other way around.

“We are very much interested in how human-driven vehicles and robots can coexist,” says Daniela Rus, director of the MIT lab and a coauthor of the paper. “It’s a grand challenge for the field of autonomy and a question that’s applicable not just for robots on roads but in general, for any kind of human-machine interaction.” One day, this kind of work might be able to help humans work more smoothly with robots on, say, the factory floor or in a hospital room.

But first, game theory. The research pulls from an approach being applied more frequently in robotics and machine learning: using games to “teach” machines to make decisions with imperfect knowledge. Game players—like drivers—often have to reach conclusions without full understanding of what the other players—or drivers—are doing. So more researchers are applying game theory to train self-driving cars how to act in uncertain situations.


Still, the uncertainty is a challenge. “Ultimately, one of the challenges of self-driving is that you’re trying to predict human behavior, and human behavior tends to not fall into rational agent models we have for game players,” says Matthew Johnson-Roberson, assistant professor of engineering at the University of Michigan and the cofounder of Refraction AI, a startup building autonomous delivery vehicles. Someone might look like they’re about to merge but see a flash of something out of the corner of their eye and stop short. It’s very hard to teach a robot to predict that kind of behavior.

Of course, driving situations could become less uncertain if the researchers were able to collect more information about human driving behavior, which is what they’re hoping to do next. Data on the speed of vehicles, where they are heading, the angle at which they’re traveling, how their position changes over time—all could help traveling robots better understand how the human mind (and personality) operates. Perhaps, the researchers say, an algorithm derived from more precise data could improve predictions about human driving behavior by 50 percent instead of 25 percent.

That might be really hard, says Johnson-Roberson. “One of the reasons I think it's going to be challenging to deploy [autonomous vehicles] is because you’re going to have to get these predictions right when traveling at high speeds in dense urban areas,” he says. Being able to tell whether a driver is a selfish driver within two seconds of observation is useful, but a car traveling at 25 mph travels nearly 75 feet in that time. A lot of unfortunate things can happen in 75 feet.

The fact is, even humans don’t understand humans all the time. “People are just the way they are, and sometimes they’re not focused on driving, and make decisions we can’t completely explain,” says Wilko Schwarting, an MIT graduate student who led the research. Good luck out there, robots.

Related Articles

Latest Articles