Papers

33

Total Citations

1,032

H-Index

16

About

Zhongyu Li is a robotics researcher whose work spans legged locomotion, safety-critical control, and human-robot interaction, with a particular focus on applying deep reinforcement learning to create agile, robust controllers for bipedal and quadrupedal robots. His most influential contributions lie in developing RL-based frameworks that enable bipedal robots—most notably the Cassie platform—to perform a remarkable range of dynamic behaviors, from versatile walking and running to precise jumping maneuvers, accumulating over 200 and 101 citations respectively for his foundational locomotion studies. Li has also pioneered adaptive control strategies, demonstrating how rapid motor adaptation techniques originally developed for quadrupeds can be successfully transferred to the far more challenging domain of bipedal systems. Beyond locomotion, his research addresses safety-critical robot operation through novel Control Barrier Function formulations, ensuring feasibility and guaranteed constraint satisfaction in real-world deployments. His creative applied work includes a leash-guided robotic guide dog for visually impaired users, quadrupedal soccer goalkeeping, and multi-robot cable-towed load transportation. With over 790 total citations across a focused body of work, Li has established himself as a leading voice in bridging theoretical reinforcement learning with practical, deployable robotic systems.

Research Focus

Key Achievements

16
H-Index
33
Papers
1,032
Total Citations
31
Avg Citations/Paper
🏆 Most Cited Paper
Reinforcement Learning for Robust Parameterized Locomotion Control of Bipedal Robots
200 citations · 2021
📈 Most Prolific Year: 2021 (10 Papers)
🤝 Key Collaborators: 67
🏛 Institutions: University of California, Berkeley, Berkeley College, Carnegie Mellon University

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago