Skip to main content.
Advanced search >
<< Back to previous page Print

<< Friday, October 13, 2017 >>


Remind me

Tell a friend

Add to my Google calendar (bCal)

Download to my calendar

Bookmark and ShareShare


Designing Automated Vehicles to Avoid Collisions (and Make Good Decisions When They Can’t)

Lecture | October 13 | 4 p.m. | 290 Hearst Memorial Mining Building


Chris Gerdes, Stanford University

Institute of Transportation Studies


Automated vehicles provide an unparalleled opportunity to reduce the approximately 35,000 fatalities that occur each year on US roads. With the ability to sense 360 degrees around the vehicle, avoid distraction, and react within milliseconds, automated vehicles possess some inherent advantages over human drivers when it comes to avoiding collisions. To realize this potential, however, the cars must be explicitly designed to make full use of these advantages when designing and executing maneuvers.

For inspiration, we have been studying race car drivers, who are able to routinely handle cars safely at the very limits of their handling capabilities. By working with expert drivers and measuring their performance on the track, we have developed automated vehicles capable of lapping a track in less time than a champion amateur driver and drifting through courses with a precision exceeding human capability. More importantly, these interactions with the best human drivers have helped us to reframe the control challenges associated with racing in a way that opens up new possibilities for safety on the road.

Even with driving capability at the level of the best human drivers, not all collisions are avoidable, due to laws of physics and the somewhat unpredictable actions of human road-users. Automated vehicles must be explicitly designed for these cases as well, requiring engineers to consider not only technical feasibility but also ethical frameworks for decision-making. The talk will conclude with a look at possible approaches to handling dilemma situations and why the popular “Trolley Car Problem” creates unnecessary fear and complication by asking the wrong question.

Chris Gerdes is a Professor of Mechanical Engineering and, by courtesy, of Aeronautics and Astronautics at Stanford University. His laboratory studies how cars move, how humans drive cars, and how to design future cars that work cooperatively with the driver or drive themselves. When not teaching on campus, he can often be found at the racetrack with students, instrumenting historic race cars or trying out their latest prototypes for the future. Vehicles in the lab include X1, an entirely student-built test vehicle; Shelley, an automated Audi TT-S that can lap a racetrack as quickly as an expert driver; and MARTY, an electrified DeLorean capable of controlled drifts. Chris and his team have been recognized with a number of awards including the Presidential Early Career Award for Scientists and Engineers, the Ralph Teetor award from SAE International and the Rudolf Kalman Award from the American Society of Mechanical Engineers.
From February 2016 to January 2017, Chris served as the first Chief Innovation Officer at the United States Department of Transportation. In this role, he worked with Secretary Anthony Foxx to foster the culture of innovation across the department and find ways to support transportation innovation taking place both inside and outside of government. He was part of the team that developed the Federal Automated Vehicles Policy and represented the Department on the National Science and Technology Committee Subcommittee on Machine Learning and Artificial Intelligence. He continues to serve U.S. DOT as Vice Chair of the Federal Advisory Committee on Automation in Transportation.
Chris is a co-founder of truck platooning company Peloton Technology and served as Peloton’s Principal Scientist before joining U.S. DOT.


jmarie@berkeley.edu