A multi-strategy fusion improved whale optimization algorithm for robot path planning

Abstract

To address path planning for robots in grid environments, this study develops an improved whale optimization algorithm by integrating several complementary search strategies. The proposed method combines mixed initialization, a nonlinear convergence factor, adaptive inertia weight, Lévy-flight perturbation, and an A*-based local repair strategy to improve search ability and path feasibility. Tent chaotic mapping and an A*-based reference path are used to enhance population diversity and initialization quality, while the nonlinear and adaptive mechanisms balance exploration and exploitation. Experiments on 20×20 and 30×30 grid maps compare IWOA with representative metaheuristic algorithms, improved WOA variants, and classical planners. The results show that IWOA achieves competitive performance in path length, fitness value, collision avoidance, obstacle clearance, and convergence efficiency. The ablation results further demonstrate the effectiveness of the key components.