Semantic mapping for mobile outdoor robots

In this paper we present the concept and realization of a semantic mapping system for a mobile outdoor robot. Semantic maps aim to give robots the ability to gather semantic information about their environment, to store it, represent it for the user, and to perform high-level tasks based on the semantic information. The map is build by a system integrating the combination of object classification and common-sense knowledge. We validate the proposed semantic map representation on a real-world 3D point cloud dataset. The presented classification approach achieves an overall precision about 96 %. The semantic maps result into a data structure which offers the opportunity to solve complex task settings and can be integrated onto real robotic systems.