Navigation method based on bionic hippocampus cognitive map

The invention relates to a navigation method based on bionic hippocampus cognitive map, which belongs to the field of bionic mobile robot navigation. A cognitive map belongs to a topology map, each node of the cognitive map comprises biological characteristic information, environment RGB-D information, and association information of multiple interdependent nodes. By combining with the cognitive map, a Dijkstra algorithm is used, through an analysis, the various information in each node and a target-heading physical distance are subjected to weighted sum, the algorithm is improved, and a globalpath with maximum characteristic information and shortest physical distance can be obtained. According to inflection point numbers, the node with the most abundant characteristic surrounding the inflection point is found by the local navigation as a terminal point of the local navigation, through SURF image characteristic point extraction, a translation matrix and a rotation matrix between two frames are compared, and th