Artificial Intelligence

U-M Taps AI to Make Autonomous Vehicle Testing 99% More Efficient

DBusiness Magazine
April 6, 2023
Download PDF
Photo by Bernd Dittrich on Unsplash

Photo by Bernd Dittrich on Unsplash


As more and more industries flock to artificial intelligence to solve data problems, the University of Michigan identified its own unique use case in autonomous vehicle testing. Early data shows the power of UofM’s artificial intelligence model reduced the need for autonomous vehicle testing miles by a jaw-dropping 99%.

The development of truly autonomous vehicles (AVs) has been hindered by the cost and time associated with safety testing, but a new system created at the University of Michigan in Ann Arbor shows that artificial intelligence can reduce the testing miles required by 99.99 percent.

This system could lead to a paradigm shift that enables manufacturers to verify more quickly whether their autonomous vehicle technology can operate safely and reduce the incidence of crashes. In a simulated environment, vehicles trained by artificial intelligence perform perilous maneuvers, forcing the AV to make decisions that confront drivers only rarely on the road, a necessary step in training the vehicles.

Read more here.

DBusiness Magazine
DBusiness Magazine

DBusiness is Detroit's premier business journal, and is part of the Hour Media family of publications. DBusiness currently has more than 31,000 subscribers and is sold at over 300 local news stands.

Become a Member