Finite-time distributed time-varying optimization of disturbed multi-agent systems
ZHU Wen-bo
ZHANG Shi-zheng
WANG Qing-ling
Abstract：Finite-time distributed optimization of multi-agent systems(MASs)refers that each agent of MASs can reach consensus at optimal state within finite-time,where optimal state is the unique minimizer of the sum of cost functions,and each cost function can be only accessed by local agent.In many practical scenarios,the cost function related to own state of agent is generally time-varying.Therefore,the distributed time-varying optimization algorithms are more desirable.Most of the existing distributed time-varying optimization algorithms for MASs are with asymptotic convergence,and the graphs are generally required to be undirected or weight-balanced,which may bring obstacles in the actual deployments.Therefore,in this paper,we study the finite-time distributed time-varying optimization problems for disturbed MASs under arbitrary strongly connected directed graph.Firstly,based on finite-time stability theory and graph theory,we propose a new class of finite-time distributed optimization algorithms which can realize distributed time-varying optimization of disturbed MASs under arbitrary strongly connected directed graphs.In addition,we extend the new distributed optimization algorithms to address the finite-time distributed time-varying optimization problems for disturbed MASs under switched directed graphs.Both of the proposed distributed algorithms can suppress the interference in multi-agent systems without using the Hessian information and the partial derivatives information of gradient with respect to time.Finally,the simulation examples are given to verify the correctness of the theoretical results.
Keywords：distributed time-varying optimizationdisturbed multi-agent systemsfinite-time stabilitydirected graphs
Publication Date：2025-03-31
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：9( 433-441 )