PSOGOOSE: A Hybrid Algorithm for UAV Path Planning in Logistics

Unmanned Aerial Vehicles (UAVs) are increasingly utilized in logistics due to their rapid deployment capabilities and flexibility. Nevertheless, constructing optimal flight paths remains a non-trivial challenge, as it requires simultaneously addressing obstacle avoidance, trajectory smoothness, and responsiveness to dynamic environments in real time. This paper presents a novel hybrid algorithm, PSOGOOSE, which integrates Particle Swarm Optimization with the Goose algorithm, further enhanced by Cauchy mutation to improve exploration and a greedy strategy to accelerate convergence. The UAV paths are expressed as spline-interpolated trajectories defined by ordered waypoints on a two-dimensional grid Simulation results demonstrate that PSOGOOSE outperforms existing methods in terms of cost, path length, constraint satisfaction, smoothness, and computational efficiency. ablation experiments and convergence analyses substantiate the robustness of the proposed approach, confirming its suitability for logistics-oriented UAV path planning tasks.