Intelligent agent layered path planning method and system considering dynamic obstacles in complex environment

The invention discloses an intelligent agent layered path planning method and system considering dynamic obstacles in a complex environment. The method comprises the following steps: constructing an environment map model of path planning; planning an optimal path of a global layer of the intelligent agent by using an A * algorithm improved by fusing environmental information and a JPS strategy; according to the global optimal path, key point extraction is carried out by using a vector method; according to the extracted key path points, a TEB method is used for optimization; moving according to the set local planning layer, performing obstacle detection in the moving process, and when encountering an obstacle, planning a local path by using a dynamic window method based on global information improvement; and carrying out loop iteration on the steps until the end point of the global path is reached. According to the invention, the problems of long time consumption, more turning points, unsmooth path and incapab