A Time-Varying Adaptive Inertia Weight based Modified PSO Algorithm for UAV Path Planning
Golam Moktader Nayeem, Mingyu Fan, Yasmin Akhter
Abstract
Path planning is an integral part of the execution of an autonomous Unmanned Ariel Vehicle (UAV). Finding the optimum path is an NP-hard problem and metaheuristics algorithms have been showing promising results in finding the optimum path. Particle Swarm optimization (PSO) is one of the commonly used metaheuristic optimization algorithms for path planning. However, PSO suffers from limitations such as falls for local minima. Particle diversity plays an important role in generating better results in path planning while avoiding local minima. Besides, parameters such as inertia weight are added to increase the diversity of particles. In this paper, we have provided an analysis of various inertia weight proposed for PSO to improve the particle diversity. Then we proposed a time-varying adaptive inertia weight parameter for our previously proposed version of PSO called nPSO for UAV path planning and compared the performance to other inertia weight parameter strategies.