Papers
7
Total Citations
105
H-Index
4
About
Lei Li is a researcher working at the intersection of robotics, artificial intelligence, and natural language processing, with particular expertise in robotic manipulation, autonomous assembly, shared control systems, and multilingual AI applications. His most impactful contribution, "Simultaneous Semantic and Collision Learning for 6-DoF Grasp Pose Estimation" (2021), has garnered 66 citations and represents a significant advance in robotic grasping, introducing an end-to-end framework that jointly learns semantic scene understanding and collision avoidance — eliminating the need for pre-known object geometry or cumbersome multi-stage pipelines. Li has also made notable strides in autonomous construction, exploring how robots can design and build bridges without blueprints, tackling challenges of long-horizon planning, sparse rewards, and variable materials through innovative reinforcement learning strategies. His work on shared control for robotic wheelchairs demonstrates a commitment to human-centered robotics, improving assistive navigation through intention prediction. Beyond robotics, Li contributed to the development of Xiaomingbot, a multilingual, multimodal news-reporting robot presented at ACL 2020. Across these diverse domains, his research consistently pushes the boundaries of intelligent, adaptable robotic systems with real-world relevance.
Research Focus
Key Achievements
Top Papers
- 1Simultaneous Semantic and Collision Learning for 6-DoF Grasp Pose Estimation66 citations · 2021
- 2
- 3Xiaomingbot: A Multilingual Robot News Reporter9 citations · 2020
- 4Learning to Design and Construct Bridge without Blueprint7 citations · 2021
- 5
- 6Xiaomingbot: A Multilingual Robot News Reporter3 citations · 2020
- 7