Papers
13
Total Citations
645
H-Index
8
About
Zipeng Fu is a robotics researcher whose work spans legged locomotion, mobile manipulation, and embodied intelligence, with a particular focus on enabling robots to operate robustly in real-world, unstructured environments. His most celebrated contribution, the Rapid Motor Adaptation (RMA) algorithm (2021, 447 citations), revolutionized quadruped robotics by allowing legged robots to adapt in real-time to unseen terrain, shifting payloads, and mechanical wear — a landmark achievement bridging simulation and physical deployment. Fu has consistently pushed the boundaries of what legged systems can achieve, from demonstrating that energy minimization naturally gives rise to emergent gaits, to developing VP-Nav, a vision-proprioception fusion system for complex navigation. His work on whole-body control unified manipulation and locomotion in legged manipulators, while Mobile ALOHA brought affordable bimanual mobile manipulation to life through imitation learning. More recently, Fu has extended his research into humanoid robotics with HumanPlus and advanced vision-language-action models through CoT-VLA. Across more than 600 cumulative citations, his research consistently translates sophisticated theoretical insights into tangible, deployable robotic systems — making him a defining voice in next-generation embodied AI.
Research Focus
Key Achievements
Top Papers
- 1RMA: Rapid Motor Adaptation for Legged Robots447 citations · 2021
- 2
- 3Coupling Vision and Proprioception for Navigation of Legged Robots37 citations · 2022
- 4
- 5
- 6CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models23 citations · 2025
- 7RMA: Rapid Motor Adaptation for Legged Robots16 citations · 2021
- 8Robot Parkour Learning10 citations · 2023
- 9Coupling Vision and Proprioception for Navigation of Legged Robots6 citations · 2022
- 10HumanPlus: Humanoid Shadowing and Imitation from Humans5 citations · 2024