Artificial Intelligence May Need a Co-pilot for Self-driving Cars
February 06, 2015A computer science professor from the University of Massachusetts is exploring how artificial intelligence can help make driving safer.
At the recent Association for the Advancement of Artificial Intelligence (AAAI) conference, Shlomo Zilberstein said that before the technology reaches full autonomy, there may be a long period where humans act as “co-pilots” or supervisors, giving some responsibility to the vehicle when possible but taking the wheel if different situations arise.
“People are unpredictable. What happens if the person is not doing what they're asked or expected to do and the car is moving at 60 miles per hour?" Zilberstein inquires. "This requires 'fault-tolerant planning.' It's the kind of planning that can handle a certain number of deviations or errors by the person who is asked to execute the plan."
The National Science Foundation has been funding Zilberstein’s efforts to study the effects of human behavior that would need to be taken into account, when preparing a robot to work semi-autonomously. He translates those ideas into a computer program, which lets a robot or autonomous vehicle plan its actions and then create a back-up plan in the event of an emergency.
Question or comment on this article? Contact an editor: engineering360editors@ihs.com
More Resources:
