Research on Improved A* Path Planning Algorithm Based on Local Environment

Path planning is a key technology to realize autonomous navigation of mobile robots and intelligent vehicles, and how to make the robot search for an optimal global path in a complex and changing environment is a current technical challenge. In this paper, based on the classical A* algorithm, we firstly establish a hybrid neighborhood search switching mechanism to make the neighborhood traversal more flexible and reasonable, and then propose a new distance estimation model based on local environment and introduce an environmental gain to adjust the heuristic function weights adaptively, and finally use quasi-uniform B-spline curve to further optimize the generated path. The experimental data and results show that the overall performance of the improved A* algorithm is better than the original algorithm. The path length is reduced by more than 10% while that of the number of traversed nodes even reaches 70%, and the path smoothness is greatly improved and the overall optimization effect is remarkable.