网络攻击下多智能体系统动态事件触发一致性  

Dynamic event-triggered consensus of multi-agent systems under cyber-attacks

作　　者：王柳 胡爱花[1] 江正仙[1] WANG Liu;HU Ai-hua;JIANG Zheng-xian(School of Science,Jiangnan University,Wuxi 214122,China)

机构地区：[1]江南大学理学院,江苏无锡214122

出　　处：《控制与决策》2023年第5期1295-1302,共8页Control and Decision

基　　金：江苏省自然科学基金项目(BK20181342);国家自然科学基金项目(61807016)。

摘　　要：研究网络攻击下一般线性多智能体系统的动态事件触发一致性问题.考虑多智能体系统在受到网络攻击后,被攻击节点的状态会改变,导致与其相应的连接无法工作,设计修复策略恢复被攻击节点及其相应的连接,给出网络攻击下分布式事件触发控制协议.在静态事件触发机制基础上,通过引入动态阈值参数,提出动态事件触发机制.进一步,利用图论、线性矩阵不等式和李雅普诺夫函数方法,给出网络攻击下实现多智能体系统一致性的充分条件,并证明在所提出的动态事件触发条件下,能够有效避免芝诺行为.最后,通过仿真例子来验证理论结果的有效性.This paper studies the dynamic event-triggered consensus of general linear multi-agent systems under cyber-attacks.Considering that the states of the attacked nodes will change after multi-agent systems are affected by cyber-attacks,which results in the failure of their corresponding connections,some repair strategies are designed to restore the attacked nodes and their corresponding connections,and a distributed event-triggered control protocol is given.Based on the static event-triggered mechanism,a dynamic event-triggered mechanism is proposed by introducing dynamic threshold parameters.Furthermore,using the graph theory,linear matrix inequality and Lyapunov function method,sufficient conditions for realizing the consensus of multi-agent systems under cyber-attacks are given,and it is proved that Zeno behavior can be effectively avoided under the proposed dynamic event-triggered conditions.Finally,a simulation example is supplied to verify the effectiveness of the theoretical results.

关 键 词：多智能体系统 网络攻击 动态事件触发机制 一致性 芝诺行为 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]