William D. Compton
William D. Compton is a doctoral student in Control and Dynamical Systems at the California Institute of Technology, where he conducts research in the AMBER Lab. His work focuses on combining reinforcement learning, nonlinear control theory, and real-time systems for legged and humanoid robotics, with an emphasis on safety-critical control and sim-to-real transfer. He has published on topics including zero dynamics policies, dynamic tube MPC, and perceptive locomotion for humanoid navigation.
“Generate, Track, Improve: Perceptive Multi-Skill Humanoid Locomotion with RL-Fine-Tuned Motion Generators”
Source→“William D. Compton, Caltech, Co-author who also contributed to the related 'Terrain Consistent Reference-Guided RL for Humanoid Navigation Autonomy'”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.