A Fusion Algorithm of Robot Path Planning Based on Improved Gray Wolf Algorithm and Dynamic Window Approach

To tackle the problems of sluggish convergence rates, low precision, and limited applicability to static settings, a fusion algorithm of the improved Gray Wolf Algorithm and the improved Dynamic Window Approach is proposed. With regard to the Gray Wolf Algorithm, an adjustable nonlinear convergence factor is used to search and optimize the balance of the algorithm in the initial phase of the iteration process, and the candidate wolves are accepted through a simulated annealing operation in the late stage of iteration to prevent local optima and achieve the global path planning. With regard to the Dynamic Window Approach, an adaptive weight is introduced to safely control the speed for passing obstacles and an offset evaluation function is enrolled to prevent excessive local path deviation. Then, the algorithm extracts key waypoints from the global path for detailed local planning, so that the initial orientation angle succeeds the orientation angle that attained the local target point on the previous occasion and generate the shortest and smooth path from the start point to the target point. Finally, the algorithm is applied to various test scenarios for path planning simulation tests, demonstrating that the proposed algorithm is feasible and possesses superior exploration capabilities.