Robust exploration and homing for autonomous robots

The ability to explore an unknown environment is an important prerequisite for building truly autonomous robots. Two central capabilities for autonomous exploration are the selection of the next view point(s) for gathering new observations and robust navigation. In this paper, we propose a novel exploration strategy that exploits background knowledge by considering previously seen environments to make better exploration decisions. We furthermore combine this approach with robust homing so that the robot can navigate back to its starting location even if the mapping system fails and does not produce a consistent map. We implemented the proposed approach in ROS and thoroughly evaluated it. The experiments indicate that our method improves the ability of a robot to explore challenging environments as well as the quality of the resulting maps. Furthermore, the robot is able to navigate back home, even if it cannot rely on its map.