Artificial Intelligence and Machine Learning

Assistance Systems Help Out Distracted Drivers

Professor Denis Gingras
Denis Gingras is teaching your car to be a better navigator by improving how it relates to its sensors, range finders, and information from other cars on the road.

Computers are taking over your car. But don’t worry: it’s not some part of a malevolent plan to get rid of their flesh-and-blood overlords. All they want to do is make you a better driver.

Unlike your fellow commuters — who might offer to help improve your driving skills through rude hand gestures and comments that can’t be reprinted in this publication — these machines offer obstacle alerts, navigation advice, or assistance when changing lanes or parallel parking. They’re called Advanced Driver Assistance Systems (ADAS) and you typically encounter them when your shiny, new car comes with built-in GPS, sensors that alert you to rearview obstacles, or even automatic braking systems to prevent collisions.

However, today’s ADAS features, increasingly prevalent in luxury cars, are just the tip of the iceberg. Eventually, your car will simply drive itself.

At least, that’s the opinion of people like Denis Gingras, a professor of engineering at Université de Sherbrooke in Quebec, Canada, and research coordinator and founding member at AUTO21, a research network centered on the auto industry. Gingras is currently working on a project for the group that aims to improve the data your car’s computer receives from its systems and other cars on the road through electronic communication. With the right combination of data, navigation and positioning can be improved — and road disasters can be automatically averted.

"The main goal is to avoid collisions altogether, instead of minimizing the risk of collisions," says Gingras.

Luxury car makers like Mercedes-Benz already use variations of ADAS position and navigation technology in some newer models. These systems trigger braking functions when the on-board computer detects the possibility of a collision. As these systems become more ubiquitous and advanced, they could have big implications for road safety, says Gingras.

"With an automated system, the computers are much faster. So you can stop the car in an emergency situation in anywhere from, say, 10 to a few hundred milliseconds," says Gingras. "You may have a gain of a factor of ten or more between the response time of an automated system and the response time of a human being. That can be the difference between a collision and no collision."

Improvements like those can’t come soon enough. While traffic deaths have fallen dramatically in the U.S. in recent decades, the National Highway Traffic Safety Administration (NHTSA) reports that more than 3,000 people died in 2010 due to "distracted-affected crashes" — or drivers who weren’t paying full attention to the road. Distraction alone accounted for over 9 percent of that year’s highway deaths.

As research progresses, those deaths might be eliminated entirely. Gingras says the move is towards fully automated cars on the mass market. Web giant Google already has working models of fully automated vehicles and ADAS capabilities are swiftly improving.

And Gingras sees no end in sight.

"The vehicles are becoming smarter. Eventually the vehicle will have its own consciousness," Gingras says. "It will be able to communicate [with other cars] more and more. It will be able to use sensors to develop a consciousness about its local environment. It will be conscious of its own position. It’s similar to human beings: we have eyes, we have ears, we talk."

Logan Kugler is a freelance technology writer based in Silicon Valley. He has written for over 60 major publications.


Join the Discussion (0)

Become a Member or Sign In to Post a Comment

The Latest from CACM

Shape the Future of Computing

ACM encourages its members to take a direct hand in shaping the future of the association. There are more ways than ever to get involved.

Get Involved

Communications of the ACM (CACM) is now a fully Open Access publication.

By opening CACM to the world, we hope to increase engagement among the broader computer science community and encourage non-members to discover the rich resources ACM has to offer.

Learn More