报告题目：Intelligent Control for Multi-agent Systems

报告人：Peng Shi 院士、教授

报告时间：2022年6月16日19:30-20:30

报告地点：腾讯会议：347-709-162

报告对象：感兴趣的教师、研究生、本科生

主办单位：beat365

报告人简介：

石碰，澳大利亚阿德莱德大学的电子与电气工程学院教授，欧洲科学院院士。获得澳大利亚纽卡斯尔大学电气工程博士学位和南澳大学数学博士学位。他的研究兴趣包括系统和控制理论, 及其在网络系统、机器人和自主系统、信息物理系统和智能系统中的应用。从2014年到2021年，他被Clarivate Analytics/Thomson Reuters (科睿唯安/汤森路透)连续评为工程和计算机科学领域的高被引科学家，连续（2019-2021）被THE AUSTRALIAN(澳大利亚人报)授予领域领导者的荣誉。他曾担任多个期刊的编委，包括Automatica和IEEE Transactions on (Automatic Control, Circuits and Systems, Fuzzy Systems)和IEEE Control Systems Letters。现任IEEE Transactions on Cybernetics的主编, Australian Journal of Electrical and Electronics Engineering的联合主编。他还担任国际系统与控制科学院经理、 IEEE SMC协会副主席和IEEE特聘讲师。 他是IEEE、IET、IEAUST、IMA和CAA的Fellow。

报告内容：

A multi-agent system (MAS) is a group of entities, referred to as agents, working together for a common goal. As the agents often work under complex circumstances, unreliable networks due to potential cyberattacks, and incomplete local information, including system uncertainty, limitations of the hardware, and unknown environment, threaten the reliability and efficiency of their collaborations. In this talk, the design of intelligent control approaches, including neural networks, reinforcement learning, and resilient control, will be presented to support resilient interaction and optimal performance of consensus and dynamic formations for MASs. A robust and distributed framework will also be introduced for harmonious collaborations of heterogenous MASs consisting of different entities. Simulation and Lab experimental results will be given to demonstrate the effectiveness of the proposed design algorithms and schemes.

欢迎全校师生踊跃参加！