Scholarlink

<p>This paper addresses the adaptive neural network (NN) event-triggered secure consensus control for stochastic nonlinear multi-agent systems (SNMASs) under sensor attacks. Distributed NNs are utilized to identify unknown nonlinear dynamics and inter-agent interaction nonlinearities, and a distributed NN state estimator is built to tackle unmeasurable agent states. Considering system spatial coupling, an NN observer is devised for each agent to distributively estimate sensor attack signals. A distributed event-triggered control (ETC) scheme is introduced to save communication resources and cut controller updates, with the triggering condition designed via local and neighboring state information. Integrating backstepping control and consensus protocols, the corresponding secure control algorithm is developed. Theoretical analysis proves all closed-loop signals are semi-globally uniformly ultimately bounded (SGUUB), and consensus tracking errors converge to a small zero neighborhood. Simulations verify the proposed scheme’s effectiveness.</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

Neural network adaptive event triggered control for stochastic nonlinear multi agent systems with sensor attacks

  • 期刊论文

  • 发表日期：

    2026年7月14日

查看全文
  • Haitong Du,
  • Shuai Sui,
  • C. L. Philip Chen

摘要

This paper addresses the adaptive neural network (NN) event-triggered secure consensus control for stochastic nonlinear multi-agent systems (SNMASs) under sensor attacks. Distributed NNs are utilized to identify unknown nonlinear dynamics and inter-agent interaction nonlinearities, and a distributed NN state estimator is built to tackle unmeasurable agent states. Considering system spatial coupling, an NN observer is devised for each agent to distributively estimate sensor attack signals. A distributed event-triggered control (ETC) scheme is introduced to save communication resources and cut controller updates, with the triggering condition designed via local and neighboring state information. Integrating backstepping control and consensus protocols, the corresponding secure control algorithm is developed. Theoretical analysis proves all closed-loop signals are semi-globally uniformly ultimately bounded (SGUUB), and consensus tracking errors converge to a small zero neighborhood. Simulations verify the proposed scheme’s effectiveness.