In more and more commercial scenarios, radio frequency identification (RFID) is used to tag assets on a large scale. These given tag infrastructures offer themselves for the navigation of autonomous transport vehicles and service robots. In this paper we investigate loop closure for graphbased simultaneous localization and mapping (SLAM) and trajectory estimation in environments with such dense RFID infrastructures: We compare different methods of inferring that a place has been revisited, examine their robustness, and show how the trajectory of the robot can be reconstructed. Given this trajectory, a robot is able to map transponder positions or to localize itself with RFID and odometry alone and without a reference localization system. The accuracy of our approach is shown through a series of experiments with a mobile robot.

Abstract. This paper addresses the problem of localization and map construction by a mobile robot in an indoor environment using only visual sensor information. Instead of trying to build high-fidelity geometric maps, we focus on constructing topological maps because they is less sensitive to poor odometry estimates and position errors. We propose a method for incrementally building topological maps for a robot which uses a panoramic camera to obtain images at various locations along its path and uses the features it tracks in the images to update the its position and the map structure. The method is very general and does not require the environment to have uniquely distinctive features. We analyze feature-based localization strategies and present experimental results in an indoor environment. 7