A hybrid topological mapping and navigation method for large area robot mapping

Abstract

In this paper, we present a hybrid topological mapping and navigation method for mobile robots. The proposed method combines metric and topological information to create map and generate navigation plan for the robot. As compared to traditional approaches of robot mapping, the method is lightweight and can be used for mapping and navigation in large areas which is particularly useful for service robots operating in large buildings. The method only uses local information for navigation while maintaining the global topological graph nodes. The topological nodes are used effectively for navigation and can also be used to store semantic information of the scene such as robot poses, scans and scene properties for complete long term robot autonomy. By combining the information from the two maps (topological and grid map), autonomous navigation and mapping in large areas for robots is possible.