Menghua Dong
Papers
6
Total Citations
151
H-Index
4
About
Menghua Dong is a robotics researcher specializing in motion planning, trajectory generation, and deep reinforcement learning (DRL) for redundant robot manipulators. His work addresses some of the most persistent challenges in robotics, particularly developing generalizable frameworks that transcend the limitations of robot-specific solutions. Dong's most influential contribution, "A General Framework of Motion Planning for Redundant Robot Manipulator Based on Deep Reinforcement Learning" (2021, 94 citations), established a foundational DRL-based approach for length-optimal path planning in obstacle-rich environments. Building on this, he has pursued inverse-kinematics-free trajectory generation methods applicable to robots with arbitrary degrees of freedom, integrating expert-guided policy learning and fuzzy feedback reward mechanisms to improve learning efficiency. His 2021 work on pose-error calibration introduced an accessible three-closed-loop transformation method that significantly reduces the technical complexity and cost of robot calibration procedures. Across his publications, Dong consistently tackles the overestimation bias in reinforcement learning, sparse reward challenges, and the generalization gap between research prototypes and real-world robotic systems. With over 150 cumulative citations, his growing body of work positions him as a meaningful contributor to intelligent robotic control, offering practical and theoretically grounded solutions for next-generation autonomous manipulation systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6