Chengxiang Liu

Guangzhou University, China University of Mining and Technology, Shenzhen University

Papers

Total Citations

103

H-Index

About

Chengxiang Liu is a pioneering robotics researcher whose work focuses on adaptive control systems, all-terrain locomotion, and multi-sensor perception for intelligent robots. His most impactful contribution, "Adaptive neural network control with optimal number of hidden nodes for trajectory tracking of robot manipulators" (46 citations), introduced a groundbreaking method for dynamically optimizing neural network architecture, significantly improving precision in robotic arm control. Liu further advanced field robotics with his design of an all-terrain wheel-legged hybrid robot (33 citations), which seamlessly integrates wheeled and legged mechanisms to overcome complex terrain obstacles—a critical innovation for search-and-rescue and exploration missions. His work on obstacle avoidance (20 citations) enhanced multi-sensor fusion by incorporating road sign detection, moving beyond traditional ultrasonic-only systems to enable context-aware navigation. Most recently, Liu developed G-SAM (4 citations), a robust one-shot keypoint detection framework that revolutionizes robot pose estimation by enabling accurate 3D positioning from single images. With a combined citation impact exceeding 100, Liu’s research bridges theoretical control algorithms and practical robotic systems, establishing him as a key figure in advancing autonomous navigation and adaptive robotics for challenging real-world environments.