Hierarchical Optimization-Based Control for Agile and Adaptive Legged Robots

Abstract:  Optimization-based control methods, including model predictive control and trajectory optimization, are essential in enhancing the capabilities of legged robots. However, deploying legged robots in real-world applications will require fast adaptation to unknown terrain, model uncertainty as well as the capability to effectively interact with unknown objects to perform practical manipulation tasks. Therefore, in this talk, I will present different directions in our research to advance optimization-based control for legged robots. Firstly, I will first discuss hierarchical model predictive control that allows legged and humanoid robots to achieve dynamic loco-manipulation. I will also introduce our work on adaptive force-based control for legged robots adapting to substantial model uncertainty and interacting with unknown objects. This work focuses on applications where legged robots traverse challenging terrains while carrying or pushing heavy loads. This talk will also discuss the combination of control, trajectory optimization and reinforcement learning toward achieving long-term adaptive behaviors in both control actions and trajectory planning for legged robots.

Bio:  Quan Nguyen is an Assistant Professor of Aerospace and Mechanical Engineering at the University of Southern California. Prior to joining USC, he was a Postdoctoral Associate in the Biomimetic Robotics Lab at the Massachusetts Institute of Technology (MIT). He received his Ph.D. from Carnegie Mellon University (CMU) in 2017 with the Best Dissertation Award. His research interests span different control and optimization approaches for highly dynamic robotics including nonlinear control, trajectory optimization, real-time optimization-based control, and robust and adaptive control. His work on the bipedal robot ATRIAS walking on stepping stones was featured on the IEEE Spectrum, TechCrunch, TechXplore and Digital Trends. His work on the MIT Cheetah 3 robot leaping on a desk was featured widely in many major media channels, including CNN, BBC, NBC, ABC, etc. Nguyen won the Best Presentation of the Session at the 2016 American Control Conference (ACC) and the Best System Paper Finalist at the 2017 Robotics: Science & Systems Conference (RSS). Nguyen is a recipient of the 2020 Charles Lee Powell Foundation Faculty Research Award.

Lab website: https://sites.usc.edu/quann/