Bootstrap learning for place recognition

Benjamin Kuipers and Patrick Beeson. Bootstrap learning for place recognition. In Proceedings of the AAAI Conference on Artificial Intelligence, pp. 174–180, Edmonton, Canada, July 2002.

Abstract

We present a method whereby a robot can learn to recognize places with high accuracy, in spite of perceptual aliasing (different places appear the same) and image variability (the same place appears differently). The first step in learning place recognition restricts attention to distinctive states identified by the map-learning algorithm, and eliminates image variability by unsupervised learning of clusters of similar sensory images. The clusters define views associated with distinctive states, often increasing perceptual aliasing. The second step eliminates perceptual aliasing by building a causal/topological map and using history information gathered during exploration to disambiguate distinctive states. The third step uses the labeled images for supervised learning of direct associations from sensory images to distinctive states. We evaluate the method using a physical mobile robot in two environments, showing high recognition rates in spite of large amounts of perceptual aliasing.

Additional Information

One of only 29 papers selected for oral presentation at AAAI 2002.

Talk slides

BibTeX

@InProceedings{Kuipers-aaai-02,
  author =       {Benjamin Kuipers and Patrick Beeson},
  title =        {Bootstrap learning for place recognition},
  booktitle =    {Proceedings of the AAAI Conference on Artificial
                  Intelligence},
  year =         2002,
  address =      {Edmonton, Canada},
  month =        {July},
  pages =        {174--180},
  abstract =     {We present a method whereby a robot can learn to recognize
                  places with high accuracy, in spite of perceptual aliasing
                  (different places appear the same) and image variability
                  (the same place appears differently). The first step in
                  learning place recognition restricts attention to
                  distinctive states identified by the map-learning algorithm,
                  and eliminates image variability by unsupervised learning of
                  clusters of similar sensory images. The clusters define
                  views associated with distinctive states, often increasing
                  perceptual aliasing. The second step eliminates perceptual
                  aliasing by building a causal/topological map and using
                  history information gathered during exploration to
                  disambiguate distinctive states. The third step uses the
                  labeled images for supervised learning of direct
                  associations from sensory images to distinctive states. We
                  evaluate the method using a physical mobile robot in two
                  environments, showing high recognition rates in spite of
                  large amounts of perceptual aliasing.},
  bib2html_pubtype ={Refereed Conference},
  bib2html_rescat ={Foundational Learning},
  bib2html_extra_info ={One of only 29 papers selected for oral presentation
                  at AAAI 2002.<br><br><a
                  href="http://personal.traclabs.com/~pbeeson/talks/Kuipers-aaai-02_talk.pdf">
                  Talk slides</a>},
}

Download:

[113.5KB pdf]