Friday, May 9, 2025 - 3:00pm
Event Calendar Category
Other LIDS Events
Speaker Name
Quan Nguyen
Affiliation
USC
Building and Room number
45-230
“Hierarchical Learning and Control for Agile and Adaptive Legged Robots”
Optimization-based control methods, such as model predictive control (MPC) and trajectory optimization, are essential to advancing the locomotion capabilities of legged robots. However, for successful deployment in real-world environments, legged robots must be capable of rapid adaptation to unknown terrain, robust handling of model uncertainties, and effective interaction with unknown objects to perform practical manipulation tasks. In this talk, I will present a hierarchical model predictive control framework that enables legged and humanoid robots to perform dynamic loco-manipulation tasks. I will also highlight our recent work on adaptive force-based control that enables legged robots to handle significant model uncertainty. These strategies allow the robots to interact robustly with unknown objects, particularly in scenarios involving the transport of heavy loads over rough terrain. Finally, I will discuss how we extend our hierarchical control principles into the domain of reinforcement learning to acquire long-term adaptive behaviors in both control actions and trajectory planning for legged and humanoid robots.
Quan Nguyen is an Assistant Professor of Aerospace and Mechanical Engineering at the University of Southern California. He is also the Chief Scientific Officer of VinMotion, a humanoid robotics company established by Vingroup, the largest private enterprise in Vietnam. 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 learning and control approaches for highly dynamic robotics, including nonlinear control, trajectory optimization, adaptive control, and reinforcement learning. His work was featured widely in many major media channels, including CNN, BBC, NBC, IEEE Spectrum, 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). He is also a recipient of the 2020 Charles Lee Powell Foundation Faculty Research Award.