Tandem drifting Toyotas show how AI might help drivers on slippery roads

Toyota Research Institute (TRI) and Stanford are plugging AI into two Supras that pull off Formula Drift-style tandem driving — but they’re looking for something more important than style points.

In a press release, TRI’s VP of human interactive driving, Avinash Balachandran, says that drifting two cars in tandem autonomously is a “milestone” and has “far-reaching implications for building advanced safety systems” in future passenger vehicles.

Beyond the impressive showing, which can be seen in a video, professor Chris Gerdes, who codirects the Center for Automotive Research at Stanford, says the physics of drifting is similar to the behavior of cars on snow or ice. Balachandran adds that the tech can kick in precisely in time to manage a driver’s loss of control, just like expert drifters. The system can solve and re-solve a problem up to 50 times per second to decide what steering, throttle, and brake commands work best in the conditions.

The dueling and drifting modified GR Supras use AI that learns from each trip on the track. TRI developed the lead car’s control mechanisms, while Stanford’s School of Engineering made the AI vehicle models and algorithms for the chase car designed to copy (and not collide with) the other. The vehicles communicate through Wi-Fi and were tuned by GReddy and Toyota Racing Development. By the way, self-drifting cars aren’t new to Stanford; a group of researchers built a DeLorean with that capability in 2015.