Wandi Wei
Papers
3
Total Citations
42
H-Index
3
About
Wandi Wei is a rising leader in humanoid robotics, specializing in the intersection of reinforcement learning and model-based control for bipedal locomotion. Her work addresses one of robotics’ grand challenges: enabling humanoid robots to walk with natural, robust gaits across diverse terrains. In her highly cited 2022 paper, Wei pioneered a hybrid approach that fuses reinforcement learning with heuristic methods, achieving agile locomotion that balances the robustness of model-based controllers with the generalization capabilities of learning-based policies. She further advanced the field by introducing motor adaptation techniques for gait-conditioned locomotion, enabling whole-body coordination without relying on perfect state estimation. Her 2024 study provides critical analytical insights into how estimation errors affect learned locomotion policies—a previously underexplored area. With over 42 citations across her key publications, Wei’s research is foundational for the next generation of humanoid robots. Her work not only pushes the boundaries of dynamic walking but also offers practical frameworks for deploying robots in real-world environments, from uneven terrain to disaster response scenarios.
Research Focus
Key Achievements
Top Papers
- 1Hybrid Bipedal Locomotion Based on Reinforcement Learning and Heuristics18 citations · 2022
- 2
- 3Toward Understanding Key Estimation in Learning Robust Humanoid Locomotion10 citations · 2024