Resilient Distributed Event-Triggered Formation Control for Multi-UAV Systems Under DoS Attacks

Maintaining the formation of multiple uncrewed aerial vehicles (UAVs) under directional denial-of-service (DoS) attacks is challenging because selective link disruptions degrade estimation, coordination, and communication efficiency. This study develops a resilient distributed framework comprising four main components. First, an adaptive topology recovery mechanism exploits the residual connectivity via confidence-weighted exponential recovery, yielding graceful degradation and smooth restoration during attack transitions. Second, a locally implementable distributed observer with locally verifiable stability margins removes dependence on global spectral information and guarantees convergence under intermittent connectivity. Third, an average-based event-triggered controller with separate position and velocity gains reduces communication while preserving tracking accuracy and enforcing a strictly positive minimum interevent time compatible with hardware limits. Fourth, a neighborhood level parameter tuning strategy uses an$\varepsilon$-constraint formulation with closed form feasible regions. We evaluate the method in 30-run Monte Carlo simulations under directional DoS attacks, and further test it under saturation-level global denial-of-service attacks against a recent baseline. Across both scenarios, the proposed method exhibits superior robustness and steady-state accuracy under the same bandwidth budget, with 30-run Monte Carlo statistics for directional DoS confirming consistent performance. The framework provides a principled solution for resilient formation control in contested communication environments.