Active Compliance for Robust Manipulation
Date: January 17, 2025 @ 3:00-3:30PM | Location: Gates B03 | Speaker: Yifan Hou | Affiliation: Stanford
Abstract:
Compliance is a physical property of motion that describes the elastic relationship brings force and motion variations. A suitable compliance profile brings robustness to robotic manipulation by handling uncertainties gracefully. In this talk, I will introduce two sets of methods for designing compliance control in manipulation tasks. I will first walk through manipulation robustness analytically, and show the role compliance control can play to improve it. With basic modeling information, the optimal control/motion plan can be computed efficiently. Then I will talk about how to learn a compliant manipulation policy directly from human demonstrations. We propose Adaptive Compliance Policy (ACP), a framework that learns to dynamically adjust system compliance both spatially and temporally for given manipulation tasks from human demonstrations.
Bio:
Dr. Yifan Hou is a Postdoctoral researcher at Stanford University, working with Prof. Shuran Song. Dr. Hou obtained his PhD and MS degrees from the Robotics Institute at Carnegie Mellon University, advised by Prof. Matthew T. Mason. Prior to joining Stanford, he was an Applied Scientist at Amazon Robotics working on the stow project. He had also spent time at Toyota Research Institute and MIT. Dr. Hou's research focuses on robotic manipulation, where he is interested in scaling up the acquisition of robust, general manipulation skills.