Event Location: GHC 4401 Rashid Auditorium
Bio: Chris Urmson is the Tech Lead of the Chauffeur team and an assistant research professor at Carnegie Mellon University (on leave). Chris was the Director of Technology for Tartan Racing, the winner of the 2007 DARPA Urban Challenge. He earned his PhD in 2005 from the Robotics Institute and his B.Sc. in Computer Engineering from the University of Manitoba in 1998. Some of his more notable claims to fame include his having been part of teams that inverted two autonomous Hummers and having been bleeped on the History Channel.

Abstract: Self-driving vehicles hold the promise of transforming the automotive industry and reshaping our relationship with the automobile. The Google self-driving car project was created to advance autonomous driving technology as quickly as possible. Building from the foundation of decades of research and the DARPA Challenges, we have developed a small fleet of autonomous vehicles.

In this talk, I will provide an overview of the work Google has been doing in advancing the state-of-the art in autonomous vehicles. To drive autonomously, vehicles use a combination of prior map data and on-line sensing. Prior to driving autonomously, we build a high resolution model of the world. On line, the vehicles use this model to estimate their position and to help them track objects that move through the world (pedestrians, cyclists, cars, etc.). The motion planner then combines an a priori model of the world with objects that are detected and tracked online to determine safe trajectories through the world.

In the course of the talk, I will demonstrate the capabilities (and limitations) of our vehicles, and talk briefly about the future of autonomous driving. There will, of course, be lots of video.