Yanghong Li

University of Science and Technology of China, Precision Research (United States)

Papers

Total Citations

H-Index

About

Yanghong Li is an emerging researcher specializing in robot learning, adaptive control, and intelligent manipulation systems. Their work sits at the intersection of deep reinforcement learning (DRL) and robotics, with a particular focus on enabling robots to interact robustly with uncertain and complex physical environments. Li's most significant contributions center on variable impedance control for robotic force tracking. Their most-cited work (15 citations) introduced a novel adaptive impedance control policy combining neural network feedforward controllers with DRL to address robust force tracking on unknown terrains — a critical challenge in contact-rich robotics. A closely related study extended this framework to grinding tasks involving complex geometries, incorporating formal stability analysis. Together, these papers establish Li as a key contributor to learning-based compliant robot control. Beyond force control, Li has advanced robotic grasping through imitation-reinforcement learning hybrids, demonstrating that unknown objects can be grasped with as few as one demonstration — dramatically reducing data requirements for anthropomorphic hand-arm systems. Their work on Bayesian deep reinforcement learning further addresses uncertainty quantification in sparse-reward manipulation tasks, tackling a fundamental limitation of conventional DRL approaches. With a rapidly growing citation record across publications from 2023–2025, Yanghong Li represents a promising voice in intelligent robotics research.