A Novel Algorithm for Robot Path Planning Based on Hybrid Strategy Improved Wild Horse Optimizer

Abstract

In order to solve the path planning problem of mobile robots, a novel hybrid strategy improved wild horse optimizer (HI-WHO) is proposed in this paper. The algorithm utilizes Sobol sequence to initialize the population, which ensures a uniform distribution of initial solutions in the search space. It also integrates Lévy flight strategy and nonlinear dynamic adaptive factor to balance the exploration and exploitation ability and improve the search efficiency and quality at different stages of the algorithm. In addition, the algorithm can ensure the global search capability and the ability to jump out of the local optimum by using the lens imaging opposition-learning strategy and greedy mechanism. In the simulation experiments, 20 benchmark functions are selected to verify the effectiveness of the proposed method. Finally, the improved algorithm, combined with cubic B-Spline interpolation, solves the path planning problem by establishing the mathematical model for mobile robot path planning, and the performance is evaluated by conducting simulation experiments in a simple experimental environment and a real environment with grid map. The results show that the HI-WHO has good stability and optimal comprehensive performance, showcasing its effectiveness in addressing the robot path planning problem.