Swarm Encirclement Planning Method of Fixed-Wing UAVs for Multiple Highly Maneuverable Targets

This paper proposes a swarm encirclement planning method for fixed-wing unmanned aerial vehicles (FW-UAVs) against multiple highly maneuverable targets, based on an improved particle swarm optimization (PSO) algorithm. First, an adaptive GA-PSO algorithm, termed A-GA-PSO, is developed by incorporating a restart mechanism, a genetic algorithm (GA), and simulated annealing (SA) into the original PSO, serving as the baseline for formation and path planning. Second, the target escape time is estimated to construct a sector-shaped escape-domain model from the initial conditions of the target and FWUAVs. Based on nearest assignment, the optimal formation is designed by considering coverage and travel distance. Finally, path planning is performed by integrating path length and obstacle avoidance requirements, ensuring trajectory feasibility while maximizing length consistency. Considering the motion characteristics of FW-UAVs, Dubins curves are adopted for trajectory smoothing. Numerical simulations validate the effectiveness and stability of the proposed method, particularly under unexpected scenarios such as drone failures.