Papers
12
Total Citations
845
H-Index
10
About
Lei Tai is a prominent robotics researcher whose work sits at the intersection of deep learning, reinforcement learning, and autonomous mobile robot navigation. His research has significantly advanced how robots perceive, learn, and operate in complex real-world environments — often without hand-crafted maps or explicitly programmed rules. Tai's most influential contributions include pioneering the use of deep reinforcement learning for mapless robot navigation, where robots learn continuous steering commands directly from sparse sensor inputs, and applying generative adversarial imitation learning to enable socially compliant navigation among pedestrians — work that has garnered over 186 citations. His "VR-Goggles for Robots" framework creatively addressed the sim-to-real transfer problem by adapting visual inputs at the robot's perception level, accumulating over 107 citations. Early foundational work on CNN-based exploration strategies (121 citations) helped establish deep learning as a practical tool for autonomous exploration in unknown environments. Beyond individual papers, Tai has contributed two widely referenced surveys on deep learning for robot control, guiding researchers through reinforcement and imitation learning paradigms. With a total body of work exceeding 800 citations, Lei Tai stands as a key figure shaping the future of intelligent, learning-enabled mobile robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2Mobile robots exploration through cnn-based reinforcement learning121 citations · 2016
- 3VR-Goggles for Robots: Real-to-Sim Domain Adaptation for Visual Control107 citations · 2019
- 4
- 5
- 6
- 7
- 8
- 9VR-Goggles for Robots: Real-to-sim Domain Adaptation for Visual Control16 citations · 2018
- 10