Researchers at the University of Waterloo developed an algorithm that uses machine-learning techniques to recognize distracted driving behaviors. The result is software that uses cameras and artificial intelligence to detect hand movements that deviate from normal driving behavior. If drivers are behaving too weirdly for too long, it can then warn them to stop texting (or whatever they are doing) before they kill someone.
The tech, developed by Fakhri Karray, Arief Koesdwiady, Chaojie Ou, and Safaa Bedawi, was presented at the 14th International Conference on Image Analysis and Recognition in Montreal. There are no reports of the software being snatched up by a car company yet, but as distracted driving is such a pervasive and deadly problem in modern society, it seems possible. In 2015, 3,477 died and 391,000 were injured because someone behind the wheel of a car was distracted.
This isn’t the first technological solution to distracted driving. There are apps that lock your phone when you’re behind the wheel, apps that let people rat out distracted drivers they spot on the road, a “textalyzer” tool that let’s police check to see if you’ve been texting, and Apple’s iOS 11 update includes a “Do Not Disturb While Driving” mode. Apple was hit with a class-action lawsuit earlier this year for not installing patented tech that could fight distracted driving on their iPhones.