Papers
20
Total Citations
313
H-Index
10
About
Guillaume Bellegarda is a robotics researcher whose work sits at the intersection of deep reinforcement learning, legged locomotion, and biologically inspired control. His research focuses on developing robust, adaptive controllers for quadruped robots — enabling them to run at high speeds, jump across variable terrain, and navigate complex environments with remarkable agility. His most cited work, "Robust High-Speed Running for Quadruped Robots via Deep Reinforcement Learning" (2022, 65 citations), demonstrated how learned locomotion policies could outperform conventional joint-space control approaches, setting a benchmark in the field. Bellegarda has also made significant contributions to dynamic jumping behaviors and gait transitions, drawing inspiration from animal neuroscience — particularly central pattern generators (CPGs) — to create more natural and adaptable robot movement. His 2024 work on viability-driven gait transitions and visually-guided locomotion reflects a growing sophistication in blending perception, biology, and machine learning. Notably, his ManyQuadrupeds framework advances the exciting goal of a single policy generalizing across diverse robot morphologies. With over 270 cumulative citations and contributions spanning hardware design, sensory feedback analysis, and model predictive control, Bellegarda has established himself as an influential voice in modern legged robotics research.
Research Focus
Key Achievements
Top Papers
- 1
- 2Robust quadruped jumping via deep reinforcement learning38 citations · 2024
- 3Robust Quadruped Jumping via Deep Reinforcement Learning31 citations · 2020
- 4
- 5Identifying important sensory feedback for learning locomotion skills22 citations · 2023
- 6
- 7
- 8
- 9Design and Evaluation of Skating Motions for a Dexterous Quadruped17 citations · 2018
- 10