“Implicit & Adaptive Representations of Human Interaction for Safe Multi-agent Collision Avoidance ” by Sylvia Herbert
Abstract: Reasoning about human-human and human-robot interaction is challenging at scale. In this talk I will show recent work of embedding implicit knowledge of human interaction into control barrier functions that allow for composable and scalable multi-agent navigation online. I will then discuss our recent work on how to update data-driven control barrier functions using Hamilton-Jacobi reachability analysis to provide safety guarantees. These projects are in collaboration with Professor Sicun (Sean) Gao at UCSD Computer Science and Engineering.
Bio: Dr. Sylvia Herbert is an Assistant Professor in Mechanical and Aerospace Engineering at the University of California San Diego. Prior to joining UCSD, she received her PhD in Electrical Engineering from UC Berkeley, where she studied with Professor Claire Tomlin on safe and efficient control of autonomous systems. Before that she earned her BS/MS at Drexel University in Mechanical Engineering. She is the recipient of the ONR Young Investigator Award, the UC Berkeley Chancellorโs Fellowship, NSF GRFP, UC Berkeley Outstanding Graduate Student Instructor Award, and the Berkeley EECS Demetri Angelakos Memorial Achievement Award for Altruism.