Papers
9
Total Citations
230
H-Index
5
About
Yuanpei Chen is a robotics and artificial intelligence researcher whose work sits at the intersection of reinforcement learning, multi-agent systems, and dexterous robotic manipulation. His research addresses some of the most demanding challenges in modern robotics: enabling robots to act safely, skillfully, and cooperatively in complex real-world environments. Chen's most influential contribution, "Safe Multi-Agent Reinforcement Learning for Multi-Robot Control" (2023, 120 citations), pioneered a safety-conscious framework for cooperative multi-robot systems — a critical step toward deploying autonomous robots in practical settings. Complementing this, his work on affordance learning ("RLAfford," 52 citations) advanced robots' ability to generalize manipulation skills across objects of varying shapes and functions. His research on bimanual dexterous manipulation pushes toward human-level robotic dexterity, including the remarkable challenge of teaching robots to dynamically throw and catch objects in real time. More recently, Chen has explored spiking neural networks for autonomous navigation and human-preference-guided learning for generating naturalistic robot behavior. Collectively, his publications reflect a coherent vision: building robots that are not only capable, but safe, adaptable, and increasingly human-like in their physical intelligence.
Research Focus
Key Achievements
Top Papers
- 1Safe multi-agent reinforcement learning for multi-robot control120 citations · 2023
- 2RLAfford: End-to-End Affordance Learning for Robotic Manipulation52 citations · 2023
- 3
- 4Active SLAM With Prior Topo-Metric Graph Starting At Uncertain Position9 citations · 2021
- 5Dynamic Handover: Throw and Catch with Bimanual Hands6 citations · 2023
- 6
- 7End-to-End Affordance Learning for Robotic Manipulation4 citations · 2022
- 8
- 9