Multi-UAV Cooperative Smoke-Screen Obscuration Strategy Based on LOS Geometric Occlusion Evaluation and PSO

Abstract

This paper addresses smoke screen occlusion protection tasks in multi-missile attack scenarios by establishing a universal solution framework that integrates geometric determination with intelligent optimization. First, motion models are developed for missiles, UAVs, and smoke grenades, abstracting targets as cylinders and describing missile observation paths via discretized line-of-sight (LOS) segments. Subsequently, an occlusion criterion based on the minimum distance between LOS segments and spherical smoke clouds is proposed, and the effective duration is computed via time discretization with boundary refinement. Under coordinated multiple smoke grenade conditions, the occlusion region is further modeled as the union of multiple spherical smoke clouds, and strategy optimization is achieved by maximizing cumulative occlusion time. For multi-target scaling, a Max-Min robust fairness objective is introduced to ensure stable performance of the protection strategy even under the most unfavorable attack direction. Finally, a Particle Swarm Optimization (PSO) algorithm is employed for global search of mixed continuous-discrete variables, with a feasibility repair mechanism designed to guarantee constraint satisfaction.