Kurt Verlin
No Comments

Toyota Wants to Make Autonomous Driving AI More Like Racing Pros

Decrease Font Size Increase Font Size Text Size Print This Page
Toyota GR Supra drifting with autonomous driving AI tech
Photo: Toyota

The Toyota Research Institute is working with Stanford University’s Dynamic Design Laboratory to develop autonomous driving AI using “the instincts of professional drivers.”

According to Gill Pratt, CEO of TRI, vehicle crashes occur every day that could, potentially, have been avoided if the drivers had superhuman skills and reflexes. TRI’s goal is to give self-driving AI those skills and reflexes using real racing drivers as a starting point.

Vehicle Tech: How to use Apple and Google assistants in your car

“The reality is that every driver has vulnerabilities, and to avoid a crash, drivers often need to make maneuvers that are beyond their abilities,” Pratt says. “Through this project, TRI will learn from some of the most skilled drivers in the world to develop sophisticated control algorithms that amplify human driving abilities and keep people safe.”

When faced with tricky conditions, professional racing drivers may not always drive a car the way you’d expect. For example, Toyota says that when faced with slippery roads, they may resort to controlled drifting as the best way through a turn — something that your average driver may not have the skills to do.

Professor Chris Gerdes says that at Stanford, this kind of research is nothing new. “Since 2008, our lab has taken inspiration from human race car drivers in designing algorithms that enable automated vehicles to handle most challenging emergencies,” he said.

Cool Concepts: Toyota makes a big splash at SEMA360 Show
Toyota Research Institute's GR Supra, equipped with autonomous driving AI technology
Photo: Toyota

More recently, the university’s Dynamic Design Lab published a paper titled “Opening New Dimensions: Vehicle Motion Planning and Control using Brakes while Drifting,” which showed off an electrified, automated DeLorean’s ability to demonstrate advanced drifting skills. They even nicknamed the car “MARTY,” and that’s just really cute.

TRI says the Stanford’s experimental results were able to be applied to create a proof-of-concept architecture that can control a rear-wheel-drive car in a drift using its steering, brakes, and propulsion system. Already, the institute is working to apply this architecture to vehicles like the GR Supra.

We have a feeling that Roborace manufacturers will be taking an even closer look at this autonomous driving AI tech than your average consumer…