is the Professor
of Department of Computer Science and Technology, Tsinghua University, China
He has been focused on neuro-fuzzy modeling, and control and filtering of nonlinear systems for over ten years. Supported by National Science Foundation of China, National Basic Research Program (973), National 863 High-Tech Program and Advanced Defensive Research Project, series of research achievements have been achieved in stable neuro-fuzzy adaptive control and robust control of nonlinear systems, and filtering and control of Markovian Jump Systems. Significant achievements are listed as follows:
1. Stable Neuro-Fuzzy Adaptive Control of Nonlinear Systems. He has proposed a novel design idea of neural network-variable structure control. He has also developed systematic design approaches to the sampled-data adaptive control of (rigid/flexible link) robotic manipulators and a class of nonlinear systems. Those theoretical results have been verified by experiments. Representative paper of this research, "Neural Network-based Adaptive Controller Design of Robotic Manipulators with an Observer", was awarded the first prize of the 18th Choon-Gang academic awards (Korea) in 2003.
2. Control and Filtering of Multi-Scale Nonlinear Systems and Markov Jump Systems.
A large class of practical plants, including flexible manipulators, flying wing aircrafts, catalytic reduction towers and large flexible space structures, can be described by multiple scale models with small perturbation parameters. He has proposed and proved the existence and sufficiency of approximation of multi-scale nonlinear systems using fuzzy singularly perturbed model (FSPM). He has also indicated the necessary condition of the universal approximator with a given precision. The adaptive and robust control approaches are developed based on both dependent and independent perturbation parameter solutions, which could obtain a significant control precision. Furthermore, effective methods are developed to address the saturation effect and incomplete measurable states in practical systems. I was invited to give a plenary speech in IEEE Intelligent System Design and Applications 2006 on this topic.
3. Application Researches. Since I joined Tsinghua University in 2001, He has supervised 20 master and 15 Ph.D students. Our research are related to some of the most important applications of neuro-fuzzy control methods in rigid and flexible-link manipulators, nonlinear systems, internetworked systems, space robots, etc. As of the experimental verification of our theoretical results, we have constructed two flexible-link robot test-beds and a teleoperation system for space on-orbit services. He has published 2 monographs, and filed 8 patent applications (3 authorized).