Zhanjun Hou

Papers

Total Citations

H-Index

About

Zhanjun Hou is a researcher whose work sits at the intersection of robotics, artificial intelligence, and autonomous navigation. His primary research focuses on applying reinforcement learning—particularly Q-learning—to enable mobile robots to make intelligent decisions in dynamic environments. Hou’s major contributions lie in developing neural network-based approaches to overcome the limitations of traditional tabular Q-learning, which requires prohibitive amounts of memory for state-action pairs in real-world robotics. By integrating neural networks with reinforcement learning, he has created more efficient and scalable learning systems for action selection and obstacle avoidance. His most cited work, “Application of reinforcement learning based on neural network to dynamic obstacle avoidance” (2008, 21 citations), demonstrates how behavior-based control architectures enhanced by reinforcement learning can achieve superior real-time performance and robustness compared to conventional model-based methods. This research has practical implications for autonomous mobile robots operating in unpredictable settings. With over 30 combined citations across his key publications, Hou’s work represents an important step toward more adaptive, learning-driven robotic systems that can navigate complex environments without exhaustive pre-programming.