Path planning for mobile robots based on improved sparrow search algorithm under various maps

With the advancement of automation technology, swarm intelligence algorithms are becoming increasingly crucial for mobile robot path planning. Therefore, an improved sparrow search algorithm (CSWSSA) is proposed to address the shortcomings of swarm intelligence algorithms in path planning, such as long planning time and suboptimal planned paths. Firstly, Utilizing Elite SPM Mapping to initialize CSWSSA to improve the individual of sparrow quality and diversity. Secondly, an improved sine cosine algorithm was proposed to increase the ability to search maps and balance the ability of extensive and fine search. Then, a tracking strategy and an adaptive T-distribution strategy are adopted at different position update points to further avoid the algorithm plunging into local minimum. Finally, in order to avoid sparrow individuals exceeding their boundaries, a boundary redistribution mechanism was designed. To test the optimization and optimization ability of the proposed CSWSSA for various functions, some extensive validations are performed on publicly available test functions. Furthermore, we conducted experimental verification on both simulated and real maps, and compared our proposed method with some widely used algorithms, and the results unanimously demonstrate the superiority of the proposed CSWSSA. Compared with other methods, the optimal path length and the average path length are shortened by 18.8% and 15.4% respectively. Moreover, CSWSSA has the best stability, with a value of 1.0255. The research results of this project will provide new ideas for path planning of intelligent optimization algorithms.