you've heard about autonomous driving, but what about autonomous drifting? that's the idea behind stanford university's delorean project, which, thanks to a pair of electric motors and some seriously impressive software, is able to slide itself with supreme accuracy around turns without any human intervention.
stanford shared a video of a car, which it calls marty, naturally, sliding through a one-kilometer obstacle course with two passengers—neither of which take control of the car at any point in the run. the car comes inches from barriers and seems to know exactly when to turn, apply throttle, and transition. it's mesmerizing to watch.
"the results so far are rather outstanding," said lead project engineer chris gerdes in a statement. "the stability control systems of modern cars limit the driver's control to a very narrow range of the car’s potential," he continued. "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."
the car has a custom suspension, bigger brakes, a full roll cage, and a suite of onboard computers mounted behind the seats. it uses two gps antennas to track its location within an inch and can calculate a drift route in seconds when given a course layout. here's how stanford describes the thought process behind developing the car's software:
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.