ABSTRACT
This paper addresses the control of cluster formations in multi-agent systems (MASs) affected by denial-of-service (DoS) attacks. An innovative Resilient dynamic event-triggered (RDET) transmission mechanism is proposed, integrating an auxiliary dynamic variable (ADV) within a sampled-data framework. Based on attack detection, a stochastic variable is introduced to schedule the triggering parameter between two different values according to the principles in a resilient manner: one value is used when there are no DoS attacks, and another value is used to increase the triggering frequency in the presence of attacks. Secondly, the agents are divided into multiple clusters to accomplish several formations and implement different tasks. By integrating specific formation information, the cluster formation control protocol is formulated, and the error system is modeled as a random system with two time-varying input delays, each with different bounds. Stability conditions, formulated through matrix inequalities, are derived by selecting a suitable Lyapunov-Krasovskii functional (LKF). Utilizing these stability conditions, the control gains and event-triggering parameters are systematically co-designed. Finally, numerical simulations involving multi-UAV systems are conducted to demonstrate the efficacy of the proposed control strategy.
International Journal of Robust and Nonlinear Control, EarlyView. Read More
