DMETM-Based Adaptive Secure Bipartite Containment Control for Stochastic Multiagent Systems Under Multipoint Attacks

The growing number of agents and frequent interactions in multiagent systems (MASs) increase the risk of excessive data transmission and network attacks. This article investigates intermittent adaptive resilient bipartite containment control for stochastic MASs under multipoint attacks. Multipoint attacks can compromise information in cooperation–competition networks, and stochastic disturbances with unknown statistical properties are almost surely nondifferentiable. To address these challenges and reduce the communication burden associated with high data transmission rates, a novel second-order event-triggered observer with feedforward compensation is proposed. Furthermore, an improved dynamic memory event-triggered mechanism (DMETM) is developed to reduce redundant signal transmissions by leveraging historical data. Unlike existing DMETMs, the proposed scheme eliminates the need to compute the supremum of the dynamic variable. The results demonstrate that the closed-loop system is bounded in practically mean square. The feasibility and superiority of the designed method are proved through simulation experiments.