Event-triggered fault-tolerant consensus control for UAV swarms under cyber attacks

Heterogeneous UAV swarms face simultaneous bandwidth constraints and adversarial cyber attacks, yet existing approaches address these in isolation. This paper presents an event-triggered fault-tolerant consensus framework for heterogeneous second-order multi-agent systems under concurrent DoS jamming and deception injections. Three coordinated mechanisms are integrated: agent-specific Zeno-free event-triggering thresholds based on relative disagreement signals; a continuous trust-evolution module with sliding-window anomaly accumulation, smooth trust decay and recovery, and a binary circuit breaker; and a trust-weighted control law attenuating compromised links in real time. A cross-term Lyapunov analysis proves uniform ultimate boundedness of all tracking errors under attack signals, sampling errors, noise, delays, and heterogeneous dynamics, with an explicit trust-topology margin guaranteeing positive stability margins. Simulations on a six-UAV swarm under concurrent DoS, deception, and GPS spoofing confirm a 62.8% reduction in transmissions, a false-data detection rate exceeding 90%, and bounded tracking errors consistent with the theoretical guarantees.