Self-triggering Secure Consensus Against Adversarial Attacks

The problem of secure consensus for multi-agent systems (MASs) is tackled in this study. The self-triggering strategy is designed to enable each healthy agent to estimate its next triggering step at the current triggering step. Thus, each healthy agent only needs to sense and broadcast at its triggering steps, and to monitor the latest broadcast states of their neighbors at their triggering steps. The frequent monitoring is thereby mitigated. Subsequently, a self-triggering secure consensus algorithm is developed to guarantee that the state variables of healthy agents reach consensus despite the influence of faulty agents in the network. The convergence analysis of the proposed method is conducted with graph tools and Lyapunov theory. Numerical examples are given to illustrate the superior performance of the proposed self-triggering secure consensus algorithm compared with the existing methods based on the static and dynamic event-triggering mechanisms.