Papers
87
Total Citations
3,602
H-Index
34
About
Zhijun Zhang is a prominent researcher whose work sits at the intersection of neural networks, optimization theory, and robotic systems. He is best known for pioneering advances in recurrent neural network (RNN) architectures designed to solve time-varying mathematical problems and for applying these methods to the motion planning and control of redundant robot manipulators. Zhang's most significant contributions include the development of varying-parameter convergent-differential neural networks (VP-CDNNs), which outperform traditional fixed-parameter approaches in solving time-varying convex quadratic programming problems—work that has garnered nearly 190 citations. His research on repetitive motion planning schemes for redundant robots, including drift-free acceleration-level control and dual-arm cyclic-motion generation for humanoid robots, has become foundational in the robotics community. He has also introduced finite-time recurrent neural networks for solving nonlinear equation systems, with direct applications to robot motion tracking. Across his body of work, Zhang has consistently bridged theoretical neural-dynamic methodologies with practical engineering challenges, such as eliminating joint-angle drift and managing physical parameter uncertainty through adaptive projection networks. With multiple papers surpassing 100 citations and a cumulative citation impact well exceeding 1,400, his research has significantly shaped how intelligent computational methods are applied to modern robotic control and real-time optimization.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Repetitive Motion Planning and Control of Redundant Robot Manipulators166 citations · 2013
- 4
- 5
- 6
- 7
- 8
- 9
- 10