Cheng-Wan An

Chinese Academy of Sciences, Shandong Institute of Automation

Papers

Total Citations

H-Index

About

Cheng-Wan An is a pioneering researcher in intelligent robotics and adaptive image processing, whose work bridges reinforcement learning and computer vision. His most influential contribution is the development of a mobile robot navigation system using neural Q-learning, a continuous Q-learning algorithm that has become a cornerstone in robotic control for its simplicity and robust theoretical foundation. This work, published in 2005 and garnering 28 citations, demonstrated how neural networks could enable autonomous navigation in complex environments, as exemplified by his work with the CASIA-I robot. In parallel, An advanced color image segmentation through rival penalized competitive learning (RPCL), addressing the critical challenge of automatically determining the number of main colors in an image—a problem that plagues traditional clustering methods. His 2005 paper on this topic, while less cited, showcases his innovative approach to unsupervised learning. An’s research has left a lasting impact on both robotics and image processing communities, providing foundational algorithms that continue to inspire new generations of researchers in autonomous systems and adaptive learning.