Papers
16
Total Citations
1,473
H-Index
13
About
Eric Cousineau is a robotics researcher whose work spans two transformative areas: bipedal humanoid locomotion and robot learning for manipulation. His foundational contributions to humanoid walking are anchored in hybrid zero dynamics (HZD), a framework he helped scale to high-degree-of-freedom robots through direct collocation and gait optimization techniques — work cited over 200 times and instrumental in making dynamic bipedal locomotion practically tractable. He also contributed to NASA's Valkyrie, one of the most ambitious humanoid platforms ever built, developed for the DARPA Robotics Challenge and cited nearly 300 times. His locomotion work on platforms such as NAO and DURUS demonstrated that human-inspired control methods could yield provably stable, efficient robotic gaits in real hardware settings. More recently, Cousineau has made significant strides in robot learning, co-authoring Diffusion Policy — now one of the most influential papers in robot manipulation, accumulating 338 citations since 2024 — which reframes visuomotor policy learning as a conditional denoising diffusion process. His Universal Manipulation Interface work further advances accessible, in-the-wild robot teaching. Across more than a decade of research, Cousineau has consistently bridged rigorous control theory with cutting-edge learning methods, leaving a substantial imprint on modern robotics.
Research Focus
Key Achievements
Top Papers
- 1Diffusion policy: Visuomotor policy learning via action diffusion338 citations · 2024
- 2Valkyrie: NASA's First Bipedal Humanoid Robot298 citations · 2015
- 3
- 4
- 5Dynamic Humanoid Locomotion: A Scalable Formulation for HZD Gait Optimization127 citations · 2018
- 6Realizing dynamic and efficient bipedal locomotion on the humanoid robot DURUS109 citations · 2016
- 7
- 8
- 9
- 10Model predictive control of underactuated bipedal robotic walking25 citations · 2015