UAV Path Planning with an Adaptive Hybrid PSO

Abstract

Optimized Path planning plays a significant role in enabling autonomous operation of Unmanned Aerial Vehicles (UAVs). Finding the shortest path is one of many NP-hard problems for which metaheuristic algorithms have shown promising results. Particle swarm optimization (PSO) is a frequently used metaheuristic approach for robotic path planning. PSO, however, has significant flaws like poor exploration capacity and a lack of diversity in the particles. PSO may become vulnerable to local minima due to either of these constraints. Inertia weight formulation and hybridization with other metaheuristic algorithms are two popular strategies for overcoming PSO's drawbacks. This study presents the aw-GPSO adaptive hybrid PSO algorithm, which is applied for offline 3D path planning of UAVs. To enhance the exploration capability of the PSO algorithm and improve particle diversity, we incorporate the Grey Wolf Optimization (GWO) algorithm and introduce a time-varying adaptive inertia weight parameter. To showcase the enhancements achieved, we conducted simulations of our proposed approach and conducted a comparative analysis with recent variants of PSO utilized in UAV path planning.