Zengliang Fang
Papers
2
Total Citations
5
H-Index
1
About
Zengliang Fang is a robotics researcher whose work bridges machine learning and autonomous manipulation, with a focus on enabling robots to learn complex tasks through demonstration and self-optimization. His key research areas include robot learning from demonstration (LfD), reinforcement learning, and adaptive control for robotic grasping and manipulation. Fang’s most cited work, "A robot demonstration method based on LWR and Q-learning algorithm" (2018, 4 citations), introduces a novel framework that combines locally weighted regression with Q-learning to allow a 6-DOF robot to adaptively learn and generate new actions for dynamic tasks like hitting a ball. This method reduces the need for explicit programming by enabling robots to generalize from human demonstrations. In a subsequent paper (2018, 1 citation), Fang extends this approach to grasping, employing Gaussian processes and Bayesian algorithms for self-learning control. Though early in his career, his contributions demonstrate a promising trajectory toward more intuitive and autonomous robotic systems, with potential applications in manufacturing and service robotics. His work exemplifies the growing integration of probabilistic machine learning into real-world robot control.
Research Focus
Key Achievements
Top Papers
- 1A robot demonstration method based on LWR and Q-learning algorithm4 citations · 2018
- 2