Improved Rao-Blackwellized Mapping by Adaptive Sampling and Active Loop-Closure

Abstract

Recently Rao-Blackwellized particle filters have been introduced as an effective means to solve the simultaneous localization and mapping (SLAM) problem. In such a particle filter each particle carries an individual map of the environment. Accordingly, a key question is how to reduce the number of particles. Additionally, the approach leaves open how to move the robot in order to improve the accuracy of the learned maps. In this paper we present novel solutions to both problems. First we present an efficient way to compute improved proposal distributions in the prediction step which drastically reduces the uncertainty about the robot's pose. Furthermore, we present an approach to effectively reduce the number of re-sampling steps which seriously reduces the particle depletion problem. Finally, we describe a technique allowing the robot to actively close loops during exploration. By re-entering already visited areas our algorithm reduces the localization error and this way produces more accurate maps. Experimental results carried out with mobile robots in large-scale indoor and outdoor environments illustrate the advantages of our methods over previous approaches.