Spring 2025 GRASP Seminar: Hersh Sanghvi, University of Pennsylvania, “Online Controller Adaptation with Meta-Learned Models”
February 7 @ 11:10 am - 11:45 am
This will be a hybrid event with in-person attendance in Wu and Chen and virtual attendance on Zoom.
ABSTRACT
Control tuning and adaptation present a significant challenge to the usage of robots in diverse environments. It is often nontrivial to find a single set of control parameters by hand that work well across the broad array of environments and conditions that a robot might encounter. Automated adaptation approaches must utilize prior knowledge about the system while adapting to significant domain shifts to find new control parameters quickly. In this talk, I will present our work to develop a general framework that deals with these challenges. I’ll discuss how we can train predictive models of controller performance that quickly adapt to online data and can be used as cost functions within efficient sampling-based optimization routines to find new control parameters online that maximize performance. I’ll also demonstrate how our framework can be used to adapt controllers for four diverse systems: a simulated race car, a simulated quadrupedal robot, and a simulated and physical quadrotor.
Presenter
Hersh Sanghvi - Learn More
Hersh Sanghvi is a Ph.D candidate in Computer and Information Science at the University of Pennsylvania GRASP Lab, advised by Dr. Camillo Jose Taylor. He received his B.S. in Electrical and Computer Engineering from the University of California, Berkeley. His research interests lie at the intersection of deep learning, control, and domain adaptation.