Improved Gray Wolf Optimization (GWO) Algorithm for Path Planning

Abstract

As we all known that existing optimization algorithms for path planning are easy to arrive local optimum state. An improved gray wolf optimization (GWO) algorithm is proposed for path planning application. GWO initializes the population randomly, which will cause poor population diversity. The searching mechanism of GWO will slow down the convergence rate and arrive the local optimal state. For the shortcomings of GWO, the initial population, searching mechanism and convergence factor have been improved. Firstly, the chaos mapping strategy is introduced to initialize the population for avoiding the uneven initial distribution of wolves individuals. Secondly, by using an adaptive solution to the convergence factor, the problem of convergence rate of GWO is improved. Finally, the weighted algorithm is presented to update the individual positions. In the paper, six classical functions are used to simulate and test particle swarm optimization (PSO), GWO and improved GWO algorithm. The results show that the improved GWO has better optimization ability and stability. The improved GWO is utilized to the path planning application with 3D non-grid map scene. The simulation results show that the proposed improved GWO for path planning achieve better performance.