Stanford engineers have taught a driverless DeLorean to drift through a race course with the precision of a human driver

Engineers in Stanford’s Dynamic Design Lab are teaching a driverless DeLorean to steer with the agility and precision of a human driver with a goal of improving how autonomous cars handle in hazardous conditions.

 

Stanford’s all-electric, autonomous DeLorean has taken another step forward in its driving expertise, flawlessly steering its way through a 1-kilometer obstacle course with the ease and precision of the best drivers in the world. The car, named MARTY, uses custom software to “drift,” the act of oversteering that allows the car to move forward when pointed sideways. This is typically a dangerous state for most drivers, but could allow autonomous cars to maneuver to quickly avoid obstacles in the road.

Commercial vehicles are outfitted with Electronic Stability Control systems that try to prevent cars from entering unstable states where the car begins to skid sideways, but this is where the best human drift drivers thrive. They harness this instability to maneuver the car to scorch through a narrow obstacle course without so much as grazing the barriers.

By studying the habits of professional drivers and testing those same control maneuvers in MARTY, the Stanford team has enabled the car to use a greater range of its physical limits to maintain stability through a broader range of conditions. The mathematics involved – which they have made open access – could allow autonomous systems to maneuver in emergencies with the agility of a professional drift racer.

“The results so far are rather outstanding,” said mechanical engineer Chris Gerdes, the lead on the project. “The stability control systems of modern cars limit the driver’s control to a very narrow range of the car’s potential. With MARTY we have been able to more broadly define the range of conditions in which we can safely operate, and we have the ability to stabilize the car in these unstable conditions.”

Comments are closed.