Chungang Zhuang
Papers
28
Total Citations
335
H-Index
9
About
Chungang Zhuang is a robotics and automation researcher whose work sits at the intersection of computer vision, machine learning, and industrial robot control. He is best known for his pioneering contributions to **6D pose estimation** and **robotic grasping**, particularly in challenging bin-picking scenarios involving cluttered, occluded, and textureless industrial objects. His papers on instance segmentation-based and semantic part segmentation-based pose estimation using point clouds and RGB-D imagery have each garnered 66 citations, establishing him as a leading voice in perception-driven manipulation. Zhuang has also made significant strides in robot dynamics modeling, introducing physics-informed neural network (PINN) approaches for friction-inclusive dynamics identification and semi-parametric deep learning models that overcome the limitations of traditional linearized methods. His fuzzy-based impedance control work for force tracking in unknown environments further demonstrates his breadth across compliant robot control. More recently, he has developed graph-based grasp pose generation frameworks and unified calibration-identification pipelines for industrial robots. With a growing publication record spanning deep learning, graph neural networks, and robot kinematics, Zhuang's research meaningfully advances the reliability and intelligence of next-generation industrial robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4A fuzzy-based impedance control for force tracking in unknown environment25 citations · 2022
- 5
- 6
- 7
- 83D Pose Estimation of Robot Arm with RGB Images Based on Deep Learning12 citations · 2019
- 9
- 10