Pengchao Zhang
Papers
5
Total Citations
155
H-Index
4
About
Pengchao Zhang is a computational intelligence and robotics researcher whose work bridges neural network theory with practical robotic applications. His most influential contribution lies in the development of varying-parameter recurrent neural networks for solving time-varying optimization problems, particularly quadratic minimization and quadratic programming challenges. His 2018 paper on power-type varying-parameter recurrent neural networks—earning over 100 citations—demonstrated the robustness of neural-dynamic methodologies in handling complex time-varying mathematical problems, establishing him as a notable voice in adaptive neural computation. Zhang has applied these theoretical foundations to real-world robotics challenges, including mobile robot path planning and redundant robot manipulator control. His modified rapidly-exploring random tree method combined with neural networks, cited 36 times, offers practical improvements to robot navigation efficiency and path smoothness. More recently, his varying-gain neural approaches for redundant manipulator self-motion represent a refinement of earlier neural dynamic frameworks, enabling more adaptive and efficient joint-space control. Across his body of work, Zhang consistently translates advanced optimization theory into actionable robotic systems, making his research particularly valuable for students and practitioners working at the intersection of machine learning, control theory, and autonomous robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4Research and Analysis on the Robot Trajectory Interpolation Methods4 citations · 2017
- 5