Max Robotics — Global Service Robot Directory + US Certification

About

Patrick Clary is a leading researcher in the field of bipedal locomotion, whose work bridges model-based control and deep reinforcement learning to create agile, energy-efficient robots. His major contributions center on the development of advanced control strategies for the Cassie and ATRIAS bipedal robots, demonstrating how passive compliance and spring-mass dynamics can be harnessed for robust walking and running. Clary’s most influential work, "Feedback Control For Cassie With Deep Reinforcement Learning" (188 citations), pioneered the use of deep RL to overcome the limitations of traditional linearized control, enabling more natural and adaptive locomotion. His research on "Fast Online Trajectory Optimization for the Bipedal Robot Cassie" (135 citations) further advanced multi-step motion planning, while his overview on energy-efficient locomotion (116 citations) highlighted critical principles for reducing the energy economy gap between robots and biological systems. Clary’s iterative RL design methodology and Monte-Carlo planning approaches have set new standards for developing dynamic locomotion skills, making him a key figure in the quest for robots that can move with the grace and efficiency of living creatures.