About

Zhaoyuan Gu is a robotics researcher whose work sits at the intersection of bipedal locomotion, motion planning, and machine learning, with a particular focus on making legged robots more capable, robust, and intelligent in real-world environments. His most influential contribution proposes a hierarchically integrated task and motion planning framework for bipedal robots navigating partially observable environments with dynamic obstacles, earning 29 citations and establishing him as a leading voice in safe autonomous locomotion. Gu has pioneered the application of signal temporal logic within model predictive control for bipedal systems — a first in the field — enabling robots to satisfy formal task guarantees while recovering from external perturbations. His early work on Adversarial A3C (20 citations) demonstrates a deep engagement with robust reinforcement learning, while more recent contributions span inverse reinforcement learning from demonstrations, whole-body humanoid loco-manipulation, and variable-stiffness robotic arms for safe human-robot interaction. His 2025 survey on humanoid locomotion and manipulation reflects his growing role as a synthesizer of the field. Collectively, Gu's research advances a vision of humanoid robots that are formally verifiable, physically resilient, and learning-capable — qualities essential for deploying legged robots beyond laboratory settings.

Research Focus

Key Achievements

7
H-Index
16
Papers
135
Total Citations
8
Avg Citations/Paper
🏆 Most Cited Paper
Integrated Task and Motion Planning for Safe Legged Navigation in Partially Observable Environments
29 citations · 2023
📈 Most Prolific Year: 2025 (4 Papers)
🤝 Key Collaborators: 71
🏛 Institutions: Georgia Institute of Technology, Applied Decision Science (United States), The Ohio State University, 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