Tianyang Pan
Papers
9
Total Citations
158
H-Index
5
About
Tianyang Pan is a leading researcher in robotic task and motion planning (TAMP), learning from demonstration (LfD), and perception for manipulation in challenging environments. His work bridges the gap between high-level discrete reasoning and low-level continuous control, enabling robots to autonomously solve complex, multi-step tasks. Pan’s most cited paper, “Learning Behavior Trees From Demonstration” (78 citations), democratizes robot programming by allowing non-experts to teach robots arbitrary tasks through demonstration. He has made foundational contributions to TAMP, including frameworks for multiple manipulators and robust planning that accounts for execution failures—critical for real-world deployment. His work on “GlassLoc” (27+ citations) tackles the notoriously difficult problem of grasping transparent objects in clutter, advancing perception under uncertainty. Pan also addresses multi-robot coordination, proposing safe navigation methods that integrate motion planning with control policies. With over 150 total citations, his research is shaping the future of autonomous systems that can operate reliably in homes, warehouses, and factories. His recent work on optimal grasps and placements further enhances TAMP efficiency in cluttered settings.
Research Focus
Key Achievements
Top Papers
- 1Learning Behavior Trees From Demonstration78 citations · 2019
- 2A General Task and Motion Planning Framework For Multiple Manipulators31 citations · 2021
- 3GlassLoc: Plenoptic Grasp Pose Detection in Transparent Clutter22 citations · 2019
- 4Failure is an option: Task and Motion Planning with Failing Executions7 citations · 2022
- 5
- 6GlassLoc: Plenoptic Grasp Pose Detection in Transparent Clutter5 citations · 2019
- 7Task and Motion Planning for Execution in the Real4 citations · 2024
- 8
- 9Optimal Grasps and Placements for Task and Motion Planning in Clutter3 citations · 2023