For mobile robot autonomous exploration and mapping in the unknown environment, the paper presents a mobile robot autonomous exploration method based on the dynamically simplified hybrid map under the robot operating system(ROS) framework. Firstly, to fast extract current frontier target points, a method of generating target points based on geometry rules is proposed. Then, from the perspective of the information gain and the cost of path, an improved value function is introduced to measure the candidate target points. Finally, an incremental caching theory is adopted to optimize topological nodes, and then construct the simplified hybrid map. Experimental results demonstrate that the proposed method has good navigation performance by improving the strategies of building topological map.