A Novel Measure of Uncertainty for Mobile Robot SLAM with Rao-Blackwellized Particle Filters

  • Authors:
  • J. L. Blanco;J. A. Fernández-Madrigal;J. Gonzalez

  • Affiliations:
  • Department of System Engineering and Automation, Universityof Málaga, 29071 Málaga, Spain, jlblanco,@ctima.uma.es;Department of System Engineering and Automation, Universityof Málaga, 29071 Málaga, Spain;Department of System Engineering and Automation, Universityof Málaga, 29071 Málaga, Spain

  • Venue:
  • International Journal of Robotics Research
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Rao-Blackwellized particle filters (RBPFs) are animplementation of sequential Bayesian filtering that has beensuccessfully applied to mobile robot simultaneous localization andmapping (SLAM) and exploration. Measuring the uncertainty of thedistribution estimated by a RBPF is required for tasks such asinformation gain-guided exploration or detecting loop closures innested loop environments. In this paper we propose a new measurethat takes the uncertainty in both the robot path and the map intoaccount. Our approach relies on the entropy of the expected map(EM) of the RBPF, a new variable built by integrating the maphypotheses from all of the particles. Unlike previous works thatuse the joint entropy of the RBPF for active exploration, ourproposal is better suited to detect opportunities to close loops, akey aspect to reduce the robot path uncertainty and consequently toimprove the quality of the maps being built. We provide atheoretical discussion and experimental results with real data thatsupport our claims.