Diyuan Liu
Papers
4
Total Citations
13
H-Index
2
About
Diyuan Liu is an emerging robotics researcher whose work sits at the intersection of reinforcement learning, locomotion control, and legged robot systems. Specializing in both bipedal humanoid and quadruped robots, Liu has made notable contributions to solving one of robotics' most persistent challenges: enabling robots to walk robustly and adaptively across complex, real-world terrain. His research consistently pushes beyond traditional model-based control methods, which suffer from high modeling complexity and limited generalizability, toward learning-based frameworks that offer greater flexibility and resilience. Liu's most cited work, "Learning-based locomotion control fusing multimodal perception for a bipedal humanoid robot" (2025, 7 citations), exemplifies his focus on integrating sensory information to achieve terrain-adaptive walking. His broader portfolio spans sim-to-real transfer optimization for quadruped robots, disturbance rejection in humanoid platforms, and energy-efficient bipedal gait generation. The development of the Xiao-Man bipedal robot platform, featured in multiple publications, highlights his hands-on engineering contributions alongside theoretical advances. With a growing citation record and publications spanning 2023 to 2025, Liu represents a promising voice in next-generation robot locomotion research, with clear implications for industrial, service, and rescue robotics applications.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4