基于观测器的多智能体系统抗DoS攻击协同控制

Observer-based Cooperative Consensus of Multi-agent Systems Against DoS Attacks

  • 摘要: 针对非线性多智能体系统,提出了一种状态观测器以应对拒绝服务(DoS)攻击导致的状态信息不确定性问题。首先,基于该观测器,设计一种分离领导者-跟随者误差与跟随者间误差的事件触发协议,以优化智能体间通信协调机制并使各个智能体可单独进行触发,降低控制器更新频率。随后,利用李雅普诺夫函数等方法分析所提控制方案的稳定性,推导实现跟随控制的充分条件。此外,严格证明任意连续触发事件间存在确定的正时间下界,从而证明所设计的触发条件和事件触发协议能够避免芝诺行为。最后,通过一个仿真实例证明了该控制策略的有效性与鲁棒性。

     

    Abstract: To address the issue of state information uncertainty in multi-agent systems under denial-of-service (DoS) attacks, we propose a state observer. Based on this observer, wen design an event-triggered protocol to separate the leader-follower errors from the inter-follower errors. This protocol aims to optimize the communication and coordination mechanism among agents, enable each agent to be triggered individually, and reduce the update frequency of controllers. Subsequently, we analyze the stability of the error system using Lyapunov functions and induction methods, with sufficient conditions for achieving follower control being derived. Furthermore, we theoretically guarantee the proposed triggering mechanism to preclude Zeno behavior through rigorous proof of a strictly positive minimum inter-event time. Finally, numerical simulations validate the effectiveness and robustness of the proposed control strategy.

     

/

返回文章
返回