Path Planning Based on Combinaion of Improved A-STAR Algorithm and DWA Algorithm
Xiaoxiao Li, Xiaoguang Hu, Ziqiang Wang, Zhuoqun Du
Abstract
In the path planning of grid map, the influence of node search strategy makes the traditional A-STAR algorithm shown with the defects of non theoretically optimal path in actual path planning, many infection points and many redundant points. Aiming at the deficiency of path planning by traditional algorithm, this paper proposes to eliminate the limitation of node movement direction in traditional A-STAR algorithm by extending its search neighborhood, design a process to remove the redundant sub nodes in the same direction due to neighborhood extension, and combine with velocity evaluation function of the improved DWA algorithm to avoid temporary obstacles. The simulation results show that relative to the traditional A-STAR algorithm, the improved A-STAR algorithm can have shorter path and fewer inflection points; compared with the traditional DWA algorithm, the dynamic path planning combined with the improved A-STAR algorithm and DWA algorithm has fewer nodes and smoother path, which can improve the working security and efficiency of mobile robot in simulation environment.