Papers
2
Total Citations
9
H-Index
2
About
Anqi Wu is a rising leader in the field of legged robotics, specializing in the intersection of reinforcement learning (RL), imitation learning, and humanoid locomotion. Her research tackles one of the most formidable challenges in robotics: enabling bipedal and humanoid robots to navigate highly uneven, dynamically changing real-world terrains with agility and robustness. Wu’s major contributions include pioneering frameworks that bridge the gap between simulation and reality. Her highly cited 2024 work, “Infer and Adapt,” introduces an inverse reinforcement learning approach that allows bipedal robots to learn reward functions directly from human demonstrations, dramatically improving their ability to adapt to complex environments. Building on this, her 2025 paper, “Learn to Teach,” proposes a sample-efficient privileged learning method that reduces the enormous simulation samples typically required for humanoid locomotion training, achieving impressive real-world performance. With over 9 citations across her most recent works, Wu’s impact is already evident in the robotics community. Her research promises to unlock transformative capabilities for humanoid robots in industrial and service applications, making her a researcher to watch in the next generation of intelligent, adaptive machines.
Research Focus
Key Achievements
Top Papers
- 1
- 2