Semi-autonomous Active Safety Systems
J. Karl Hedrick
University of California Berkeley, Department of Mechanical Engineering
Friday, April 19, 2013|
3:30pm - 4:30pm
Add to Google Calendar
About the Event
Over the past few decades, intelligent vehicle systems have become increasingly important in their contribution to vehicle safety. A semi-autonomous system that explicitly models the driver’s behavior is able to make accurate assumptions on the predicted safety of the driver. Rather than disconnecting the driver from the vehicle, our approach incorporates the driver using a model predictive control method to predict the future trajectories of the closed-loop driver-vehicle system. We can then add the minimal corrective action necessary to keep the driver from violating safety constraints. The successful implementation of this controller, however, is predicated on a reliable approximation of the environment and the vehicle's inertial properties. A system that can account for these variations can execute more accurate control. In addition, it is useful to model the interaction between the tires and the road to correctly predict the vehicle’s trajectory. Although most contemporary vehicle systems compute the control inputs based on a fixed tire-road friction coefficient, real-time estimation is vital for improvement of the performance of vehicle control systems. Use of in-tire accelerometers allows us to classify the road surface conditions with direct measurement from the tire. This integrated approach, combining accurate inertial parameter estimates, real-time estimation of tire-road adhesion, and a control system that explicitly models driver behavior, allows the development of a new echelon of driver assistance systems.
J. Karl Hedrick is the James Marshall Wells Professor of Mechanical Engineering at the University of California (UC) at Berkeley. He received his BS degree (1966) from the University of Michigan, and his MS (1970) and PhD (1971) degrees from Stanford University. His research focuses on the application of advanced control theory to a wide variety of vehicle dynamic systems including automotive, aircraft and ocean vehicles. He is currently the Director of Berkeley’s Vehicle Dynamics Laboratory and has served as the Director of the UC PATH Research Center (1997-2003). He has received numerous awards, including 2006 ASME Rufus Oldenburger Medal, ASME JDSMC Best Paper Award (1983&2001), IEEE TCST Outstanding Paper Award (1998), the AACC O. Hugo Schuck Best Paper Award (2003) and gave the Nyquist Lecture at the ASME 2009 DSCC.
Contact: Ann Pace
Sponsor: Bosch, Eaton, Ford, GM, Toyota, Whirlpool and the MathWorks
Open to: Public