<p>This paper discusses the consensus issue of nonlinear multi-agent systems (MASs) operating over directed communication topologies under Denial-of-Service (DoS) attacks. To overcome the challenges posed by intermittent communication failures and the asymmetry of directed networks, a resilient control scheme integrating adaptive input adjustment and an event-triggered strategy is proposed. The main contributions include the design of a dual-terminal asynchronous dynamic event-triggered (DET) mechanism, which effectively reduces communication and controller updates by coordinating measurement and control channels triggering mechanisms. Within this framework, each agent independently broadcasts its state information based on edge-dependent triggering rules, enabling fully decentralized and asynchronous communication over the directed graph. Furthermore, to maintain control performance during DoS attacks, the triggering error is replaced by the observer’s estimate of the actual state. By systematically analyzing the duration and frequency characteristics of DoS attacks, several stability conditions for nonlinear MASs under attacks are established. Finally, some examples are given to prove the effectiveness of the proposed method.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Dynamic event-triggered consensus of nonlinear multi-agent systems under DoS attacks: dual-terminal asynchronous control

  • Xiaoli Ruan,
  • Ze Tang,
  • Ailong Wu,
  • Jianwen Feng,
  • Jingyi Wang

摘要

This paper discusses the consensus issue of nonlinear multi-agent systems (MASs) operating over directed communication topologies under Denial-of-Service (DoS) attacks. To overcome the challenges posed by intermittent communication failures and the asymmetry of directed networks, a resilient control scheme integrating adaptive input adjustment and an event-triggered strategy is proposed. The main contributions include the design of a dual-terminal asynchronous dynamic event-triggered (DET) mechanism, which effectively reduces communication and controller updates by coordinating measurement and control channels triggering mechanisms. Within this framework, each agent independently broadcasts its state information based on edge-dependent triggering rules, enabling fully decentralized and asynchronous communication over the directed graph. Furthermore, to maintain control performance during DoS attacks, the triggering error is replaced by the observer’s estimate of the actual state. By systematically analyzing the duration and frequency characteristics of DoS attacks, several stability conditions for nonlinear MASs under attacks are established. Finally, some examples are given to prove the effectiveness of the proposed method.