Songyin Cao

Shenyang Ligong University, Yangzhou University

Papers

Total Citations

H-Index

About

Songyin Cao is a researcher whose work sits at the intersection of intelligent control systems and autonomous robotics, with a particular focus on enhancing the precision and reliability of robotic manipulators and perception systems. His foundational contribution, "Research on RBF neural network model compensation and adaptive control of robot manipulators" (2016, 5 citations), addresses a critical challenge in robotics: the nonlinearity and uncertainty inherent in manipulator dynamics. By integrating Radial Basis Function (RBF) neural networks with computed torque control on PD, Cao developed an adaptive control method that compensates for model errors, significantly improving trajectory accuracy. This work has provided a robust framework for controlling complex robotic systems in uncertain environments. More recently, Cao has ventured into the domain of autonomous perception with "A Dynamic Visual SLAM System based on YOLOv5s and Scene Flow" (2025, 1 citation), tackling the persistent problem of dynamic interference in SLAM systems. By leveraging YOLOv5s for object detection and scene flow for motion analysis, his approach enhances the robustness of visual SLAM in real-world, dynamic settings—a vital step for applications in autonomous driving and field robotics. Through these contributions, Cao demonstrates a clear trajectory from foundational control theory to cutting-edge perception systems, marking him as a thoughtful contributor to the evolution of intelligent robotics.