Papers
1
Total Citations
4
H-Index
1
About
Haiyang He is a rising researcher in robotics and embodied intelligence, whose work focuses on the intersection of imitation learning and reinforcement learning for dexterous manipulation. His key research areas include robotic grasping, anthropomorphic hand-arm systems, and sample-efficient learning for unknown objects. He is best known for his pioneering paper, "Grasping Unknown Objects With Only One Demonstration" (2024, 4 citations), which addresses a critical challenge in robotics: enabling robots to grasp novel objects with minimal human input. By combining imitation learning with reinforcement learning, He’s method reduces the need for extensive, perfect demonstrations—a major bottleneck in real-world deployment. This work has immediate implications for manufacturing, assistive robotics, and household automation, where robots must adapt to unfamiliar objects quickly. Though early in his career, He’s approach to bridging the simulation-to-reality gap and his focus on data efficiency mark him as a promising innovator. His research offers a practical path toward more autonomous, learning-driven robotic systems, making him a name to watch in the field of robot manipulation.
Research Focus
Key Achievements
Top Papers
- 1Grasping Unknown Objects With Only One Demonstration4 citations · 2024