文献类型：期刊文献

中文题名：Distributed hybrid optimization for multi-agent systems

英文题名：Distributed hybrid optimization for multi-agent systems

作者：Tan XueGang[1,2];Yuan Yang[2];He WangLi[2];Cao JinDe[1];Huang TingWen[3]

机构：[1]Southeast Univ, Sch Math, Nanjing 210096, Peoples R China;[2]East China Univ Sci & Technol, Minist Educ, Key Lab Smart Mfg Energy Chem Proc, Shanghai 200237, Peoples R China;[3]Texas A&M Univ Qatar, Dept Sci, Doha 23874, Qatar

年份：2022

卷号：65

期号：8

起止页码：1651

中文期刊名：Science China(Technological Sciences)

外文期刊名：SCIENCE CHINA-TECHNOLOGICAL SCIENCES

收录：CSTPCD;;EI(收录号：20223012404211);Scopus;WOS:【SCI-EXPANDED(收录号:WOS:000827912700003)】；CSCD:【CSCD2021_2022】；PubMed;

基金：This work was supported in part by the National Key Research and Development Program of China (Grant No. 2020YFA0714300), the National Natural Science Foundation of China (Grant Nos. 61833005 and 62003084), the Fundamental Research Funds for the Central Universities, the Jiangsu Provincial Key Laboratory of Networked Collective Intelligence (Grant No. BM2017002).

语种：英文

中文关键词：multi-agent systems;hybrid impulsive strategy;optimal consensus;distributed optimization

外文关键词：multi-agent systems; hybrid impulsive strategy; optimal consensus; distributed optimization

摘要：This paper addresses the distributed optimization problems of multi-agent systems using a distributed hybrid impulsive protocol.The objective is to ensure the agents achieve the state consensus and optimize the aggregate objective functions assigned for each agent with distributed manner. We establish two criteria related to the optimality condition and the impulsive gain upper estimation, and propose a distributed hybrid impulsive optimal protocol, which includes two terms: the local averaging term in the continuous interval and the term involving the gradient information at impulsive instants. The simulation results show that the optimal consensus can be realized under the distributed hybrid impulsive optimization algorithm.
This paper addresses the distributed optimization problems of multi-agent systems using a distributed hybrid impulsive protocol. The objective is to ensure the agents achieve the state consensus and optimize the aggregate objective functions assigned for each agent with distributed manner. We establish two criteria related to the optimality condition and the impulsive gain upper estimation, and propose a distributed hybrid impulsive optimal protocol, which includes two terms: the local averaging term in the continuous interval and the term involving the gradient information at impulsive instants. The simulation results show that the optimal consensus can be realized under the distributed hybrid impulsive optimization algorithm.