Communication-Aware UAV Formation Control for Cooperative Sensing

Abstract

The efficacy of unmanned aerial vehicle (UAV) swarm cooperative sensing fundamentally depends on threedimensional (3D) formation geometry, which governs target observability, estimation accuracy, and inter-node connectivity. In existing literature, formation configuration has been frequently treated as a secondary implementation detail, with Cramér-Rao Lower Bound (CRLB) optimizations yielding only abstract direction vectors that disregard practical wireless constraints and executable deployment coordinates. To bridge this critical gap, this paper proposes a communication-aware formation control framework that tightly couples information-theoretic sensing optimization under signal-to-noise ratio (SNR) constraints. First, we employ a Lyapunov-based gradient descent scheme to minimize the CRLB trace for time-of-arrival (TOA) sensing. Then, by integrating both TOA measurement and air-toair (A2A) communication SNR constraints, we map theoretical direction vectors into physically feasible 3D position intervals that simultaneously guarantee localization fidelity and communication reliability. To realize autonomous formation transitions, we develop an enhanced artificial potential field (APF) method featuring segment-guided attraction, velocity-adaptive repulsion, and a dedicated communication-aware potential field. Extensive simulations demonstrate that the proposed method achieves realtime obstacle avoidance without over-conservative detours, and maintains robust inter-UAV connectivity across varying SNR thresholds and swarm scales.