Autonomous Robots

, Volume 12, Issue 3, pp 287–300 | Cite as

Fast, On-Line Learning of Globally Consistent Maps

  • Tom Duckett
  • Stephen Marsland
  • Jonathan Shapiro


To navigate in unknown environments, mobile robots require the ability to build their own maps. A major problem for robot map building is that odometry-based dead reckoning cannot be used to assign accurate global position information to a map because of cumulative drift errors. This paper introduces a fast, on-line algorithm for learning geometrically consistent maps using only local metric information. The algorithm works by using a relaxation technique to minimize an energy function over many small steps. The approach differs from previous work in that it is computationally cheap, easy to implement and is proven to converge to a globally optimal solution. Experiments are presented in which large, complex environments were successfully mapped by a real robot.

simultaneous localization and mapping concurrent map-building and self-localization relaxation algorithm Gibbs sampling learning and adaptation 


  1. Burgard, W., Fox, D., Jans, H., Matenar, C., and Thrun, S. 1999. Sonar-based mapping with mobile robots using EM. In Proc. of the International Conference on Machine Learning.Google Scholar
  2. Duckett, T. 2000. Concurrent map building and self-localisation for mobile robot navigation. Ph.D. Thesis, Department of Computer Science, University of Manchester.Google Scholar
  3. Duckett, T., Marsland, S., and Shapiro, J. 2000. Learning globally consistent maps by relaxation. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA'2000), San Francisco, CA.Google Scholar
  4. Duckett, T. and Nehmzow, U. 1999. Exploration of unknown environments using a compass, topological map and neural network. In Proceedings of the 1999 IEEE International Symposium on Computational Intelligence in Robotics and Automation, Monterey, CA, pp. 312–317.Google Scholar
  5. Duckett, T. and Nehmzow, U. 2001. Mobile robot self-localisation using occupancy histograms and a mixture of Gaussian location hypotheses. Robotics and Autonomous Systems, 34(2/3):119–130.Google Scholar
  6. Frese, U. and Hirzinger, G. 2001. Simultaneous localization and mapping—A discussion. In Proceedings of the IJCAI-2001Workshop on Reasoning with Uncertainty in Robotics (RUR'01), D. Fox and A. Saffiotti (Eds.).Google Scholar
  7. Golfarelli, M., Maio, D., and Rizzi, S. 1998. Elastic correction of dead-reckoning errors in map building. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS'98), Victoria, Canada, pp. 905–911.Google Scholar
  8. Gutmann, J.-S. and Konolige, K. 1999. Incremental mapping of large cyclic environments. In Proceedings of the 1999 IEEE International Symposium on Computational Intelligence in Robotics and Automation, Monterey, CA.Google Scholar
  9. Hinkel, R. and Knieriemen, T. 1988. Environment perception with a laser radar in a fast moving robot. In Symposium on Robot Control (SYROCO'88), Karlsruhe, Germany.Google Scholar
  10. Hopfield, J. 1982. Neural networks and physical systems with emergent collective computational abilities. In Proceedings of the National Academy of Sciences, USA, 79:2554–2558.Google Scholar
  11. Lu, F. and Milios, E. 1997a. Globally consistent range scan alignment for environment mapping. Autonomous Robots, 4:333–349.Google Scholar
  12. Lu, F. and Milios, E. 1997b. Robot pose estimation in unknown environments by matching 2D range scans. Intelligent and Robotics Systems, 18:249–275.Google Scholar
  13. Marsland, S., Nehmzow, U., and Duckett, T. 2001. Learning to select distinctive landmarks for mobile robot navigation. Robotics and Autonomous Systems, 37(4):241–260.Google Scholar
  14. Moravec, H. and Elfes, A. 1985. High resolution maps from wide angle sonar. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA'85), St. Louis, Missouri, pp. 116–121.Google Scholar
  15. Reif, F. 1982. Fundamentals of Statistical and Thermal Physics. McGraw-Hill: New York.Google Scholar
  16. Shatkay, H. 1998. Learning models for robot navigation. Ph.D. Thesis, Department of Computer Science, Brown University.Google Scholar
  17. Shatkay, H. and Kaelbling, L.P. 1997. Learning topological maps with weak local odometric information. In Proceedings of the 15th International Joint Conference on Artificial Intelligence.Google Scholar
  18. Smith, R. and Cheeseman, P. 1986. On the representation and estimation of spatial uncertainty. International Journal of Robotics Research, 5(4):56–68.Google Scholar
  19. Thrun, S., Burgard, W., and Fox, D. 1998. Aprobabilistic approach to concurrent mapping and localisation for mobile robots. Machine Learning, 31(5):1–25.Google Scholar
  20. Weiss, G. and von Puttkamer, E. 1995. A map based on laserscans without geometric interpretation. In Proceedings of Intelligent Autonomous Systems 4 (IAS-4), Karlsruhe, Germany, pp. 403–407.Google Scholar
  21. Yamauchi, B. and Beer, R. 1996. Spatial learning for navigation in dynamic environments. IEEE Transactions on Systems, Man and Cybernetics Section B, 26(3).Google Scholar

Copyright information

© Kluwer Academic Publishers 2002

Authors and Affiliations

  • Tom Duckett
    • 1
  • Stephen Marsland
    • 2
  • Jonathan Shapiro
    • 2
  1. 1.Department of TechnologyUniversity of ÖrebroÖrebroSweden
  2. 2.Department of Computer ScienceUniversity of ManchesterManchesterUK

Personalised recommendations