Self-Driving Cars Capable of Extreme Evasive Moves
VIDEO: Self-Driving DeLorean Performs Doughnuts
Imagine riding in a self-driving car that's capable of the kind of extreme collision-avoidance maneuvers normally reserved for professional drivers who specialize in movie stunts or competitive racing.
That's a research mission taking shape at Stanford University. A team of engineers has transformed a vintage 1981 DeLorean into an electric vehicle fit to test the physical limits of autonomous driving. The latest addition to Stanford’s research fleet is nicknamed MARTY– short for Multiple Actuator Research Test bed for Yaw control. As you’ve probably guessed, the name pays homage to Michael J. Fox’s character in “Back to the Future.”
“We want to design automated vehicles that can take any action necessary to avoid an accident,” explained Chris Gerdes, a professor of mechanical engineering. “The laws of physics will limit what the car can do, but we think the software should be capable of any possible maneuver within those limits. MARTY is another step in this direction, thanks to the passion and hard work of our students.”
Playing a key role in the project is Jonathan Goh, a mechanical engineering graduate student in Gerdes’ Dynamic Design Lab (DDL). One of his challenges has been programming the car to handle all operating regimes, not just the simple one that electronic stability control imposes. The vehicle needs to be capable of determining when to sacrifice vehicle stability in order to execute a demanding evasive move.
“When you watch a pro driver drift a car, you think to yourself that this person really knows how to precisely control the path and angle of the car, despite how different it is from normal driving,” Goh said. “The wheels are pointed to the left even though the car is turning right, and you have to very quickly coordinate the throttle and steering in order to keep the car from spinning out or going the wrong way. Autonomous cars need to learn from this in order to truly be as good as the best drivers out there.”
Eventually, the test car will learn how to race around a track using this drifting technique to negotiate tight turns around obstacles when necessary. Already, the car can autonomously lock itself into a continuous, precisely circular doughnut at a large drift angle. The feat demonstrates Goh’s expertise in controls engineering.
Stanford researchers built MARTY in collaboration with Renovo Motors, a Silicon Valley company specializing in advanced electric vehicle technology. The partnership gave the Stanford team early access to a new platform derived from Renovo’s electric vehicle that delivers 4,000 pound-feet from on-motor gearboxes to the rear wheels in a fraction of a second. This permitted precise control of the forces required to drift. A central application programming interface (API) manages the systems, allowing a rapid integration process.
The collaboration has allowed Gerdes’ students to focus on the subsystems and algorithms most critical to the research project’s goals. On Oct. 20, the Revs Program at Stanford and the Center for Automotive Research at Stanford (CARS) hosted an event that introduced the vehicle to the public.
“In our work developing autonomous driving algorithms, we’ve found that sometimes you need to sacrifice stability to turn sharply and avoid accidents,” Gerdes said. “The very best rally car drivers do this all the time, sacrificing stability so they can use all of the car’s capabilities to avoid obstacles and negotiate tight turns at speed. Their confidence in their ability to control the car opens up new possibilities for the car’s motion. Current control systems designed to assist a human driver, however, don’t allow this sort of maneuvering. We think that it is important to open up this design space to develop fully automated cars that are as safe as possible.”
To view a Stanford News Service video about the project, click on the photo or link below the headline.