Qijin She

National University of Defense Technology

Papers

Total Citations

H-Index

About

Qijin She is a rising researcher in robotics and artificial intelligence, with a primary focus on dexterous manipulation and high-degree-of-freedom (DOF) robotic control. Their most notable contribution is the development of cross-hand policy learning for complex reaching and grasping tasks, as demonstrated in their 2024 paper "Learning Cross-Hand Policies of High-DOF Reaching and Grasping." This work addresses a critical challenge in robotics: enabling robots to transfer manipulation skills across different hand morphologies, such as from a three-fingered gripper to a five-fingered anthropomorphic hand. By leveraging reinforcement learning and domain randomization, She's approach allows robots to generalize grasping strategies without extensive retraining, significantly improving adaptability in unstructured environments. Though early in their career, with their flagship paper already garnering 2 citations, She's research has implications for assistive robotics, prosthetics, and industrial automation. Their work stands out for tackling the intersection of motor control, transfer learning, and high-dimensional action spaces—a promising direction for more versatile and intelligent robotic systems.