Papers
4
Total Citations
23
H-Index
3
About
Anqiao Li is a robotics researcher focused on advancing autonomous locomotion for legged and mobile ground robots. Their work centers on two key areas: robust quadrupedal gait control and semantic perception for support surface estimation. Li’s most impactful contribution is the development of a semantic pointcloud filter that enables mobile robots to accurately perceive their support surface through visual clutter like grass, addressing a critical failure mode in outdoor autonomous navigation. This work has garnered 13 citations since 2023. In quadrupedal locomotion, Li pioneered an efficient deep reinforcement learning framework that uses pretrained neural networks to learn robust bounding gaits—a dynamic gait essential for obstacle negotiation. Their approach significantly reduces training time while maintaining natural, stable motion, as demonstrated in their 2022 and 2020 publications. Li’s research bridges the gap between perception and control, offering practical solutions for robots operating in unstructured environments. Their work on pre-fitting neural networks for bound controllers further showcases their ability to combine simulation-based pretraining with real-world deployment, making their methods both scalable and applicable to field robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4