Chang Deng

Papers

Total Citations

H-Index

About

Chang Deng’s research lies at the intersection of computational intelligence, reinforcement learning, and autonomous robotics, with a particular focus on fuzzy control systems. Deng’s most significant contribution is the development of **dynamic fuzzy Q-learning (DFQL)**—a pioneering method that enables mobile robots to navigate efficiently by automatically generating and tuning fuzzy rules through Q-learning. This work, published in 2003 and cited 10 times, demonstrated how continuous-valued states and actions could be handled using fuzzy reasoning, effectively bridging the gap between reinforcement learning and fuzzy inference. In a follow-up 2004 paper (4 citations), Deng extended this framework to allow for the **online, self-organizing generation of fuzzy inference systems**, where both structure and parameter identification occur automatically and simultaneously without human intervention. These contributions are notable for their practical impact on autonomous navigation and adaptive control, offering a scalable, learning-based approach to real-time decision-making in uncertain environments. Deng’s work remains a foundational reference for researchers exploring hybrid intelligent systems and adaptive robot control.