Watch Toyota and Stanford execute a two-car autonomous drift

Toyota has reached an unusual milestone. Working with Stanford University, and leveraging artificial intelligence, it managed to make two cars autonomously drift around a track at the same time. It hopes this feat will help it design safer electronic driving aids in the coming years.

Somewhat surprisingly, both Toyota and Stanford have amassed a great deal of experience in autonomous drifting technology. Toyota built a Supra-based prototype in 2021, and Stanford taught a DeLorean DMC-12 how to drift in 2019. Sending two autonomous cars out at the same time has never been done, according to Toyota, and researchers say it’s far more difficult than it sounds. Keeping the two Supra-based test mules sideways, mere inches away from each other, without cracking plastic or bending sheetmetal required using artificial intelligence that learns the ins and outs of drifting lap after lap. The technology is capable of adapting to changing track conditions, for example.

Real-world testing took place at the Thunderhill Raceway Park in Willows, California. Each car served a different purpose: The first was set up as the lead car and the second was configured as a chase car. Toyota enlisted the help of well-known tuner GReddy to modify each car’s engine, transmission, suspension system, and safety systems, and both prototypes were fitted with a roll cage. While technical information hasn’t been released, Toyota notes that the two prototypes were built to the same specifications used in Formula Drift competitions.

An armada of onboard sensors and computers control the steering, the throttle, and the brakes during the drift, and the two cars talk to each other via a dedicated Wi-Fi network. They notably share information such as their current position and where they’re planning on going. 

Toyota and Stanford aren’t preparing to launch a drifting series for autonomous cars. They will use the lessons learned during this project to develop safer driving aids in the coming years. “The physics of drifting are actually similar to what a car might experience on snow or ice. What we have learned from this autonomous drifting project has already led to new techniques for controlling automated vehicles safely on ice,” explained Chris Gerdes, a professor of mechanical engineering and the co-direct of the Center for Automotive Research at Stanford.