Resilient Leader-Following Consensus of Multi-agent Systems Against Byzantine Attacks with a Non-autonomous Leader

This letter aims to tackle the continuous-time distributed Byzantine-resilient leader-following consensus problem in multi-agent systems. The approach is built on directed graphs that use single-order integrator. Within a brief period, a novel edge-based protocol is created to enable followers to monitor the status of the non-autonomous leader. Then it is proved by non-smooth analysis that the above protocols guarantee finite-time zero-error control for topologies with strong \((2f+1)\) robustness. By using a numerical simulation example, this edge-based protocol is illustrated as being practicable and valid.