博士论文答辩：模式识别抗干扰策略研究：基于特征的补偿
日期：2024/05/10 - 2024/05/10
博士论文答辩：模式识别抗干扰策略研究：基于特征的补偿
主讲人：胡晓波
时间：2024年5月10日（周五）下午14:00
地点: 密西根学院龙宾楼414A
讲座摘要
Pattern recognition is an important discipline of artificial intelligence. However, the recognition processes are inevitably subject to parameters perturbation, unmodeled dynamics, state coupling, and external disturbances, which lower the system performance. These instability problems bring challenges to the design and application of pattern recognition. Therefore, this dissertation develops a research on the active disturbance rejection control in pattern recognition. The unified hierarchical recognition model establishes a mathematical basis for the developments of control strategy. Then, the active disturbance rejection control is realized towards the disturbance observation and compensation in a controllable recognition system. The last but not least, the stability condition of the recognition system is theoretically analyzed. The proposed theoretical framework and control strategies in this dissertation improve the robustness of the pattern recognition system in those complex working scenarios. Through the mathematics developments and experiments validation, the application of active disturbance rejection control theory can provide a theoretical guidance for the design and application of a more robust pattern recognition system.
主讲人简介
Xiaobo Hu received the B.Eng. degree in mechanical engineering from Southern China University Technology, Guangzhou, China, in 2016, and the M.Sc. degree in mechanical engineering from Shanghai Jiao Tong University, Shanghai, China, in 2019. He is currently pursuing the Ph.D. degree with the Joint Institute of UM-SJTU, Shanghai Jiao Tong University, Shanghai, China, supervised by Prof. Jun Zhang and Jianbo Su. His current research interests include disturbance rejection control, robotics and machine learning.