Self-Organized Semantic Localization
Matt Weber, Edward A. Lee

Citation
Matt Weber, Edward A. Lee. "Self-Organized Semantic Localization". Talk or presentation, 16, May, 2014.

Abstract
Indoor localization is a hard problem for many physical reasons, and the majority of research attention has been devoted to devising mechanisms to achieve better spatial accuracy. However beyond gathering data, an equally important (but often ignored) question is what to do with location information when it is available. We propose semantic relations as building blocks for a first order language of localization. This representation may be used to express meaningful questions about the relative position of objects without necessarily resorting to physical coordinates. Since the same subset of a first order language can be used for different models when an elementary embedding exists between the backing structures, our approach does not restrict the actual representations of space provided they are so related. A heterogeneous strategy employing a combination of localization methods and different spatial representations for different areas can be accomplished by this method.

Electronic downloads

Internal. This publication has been marked by the author for TerraSwarm-only distribution, so electronic downloads are not available without logging in.

Citation formats

• HTML

  Matt Weber, Edward A. Lee. <a
  href="http://www.terraswarm.org/pubs/478.html"
  ><i>Self-Organized Semantic
  Localization</i></a>, Talk or presentation,  16,
  May, 2014.

• Plain text

  Matt Weber, Edward A. Lee. "Self-Organized Semantic
  Localization". Talk or presentation,  16, May, 2014.

• BibTeX

  @presentation{WeberLee14_SelfOrganizedSemanticLocalization,
      author = {Matt Weber and Edward A. Lee},
      title = {Self-Organized Semantic Localization},
      day = {16},
      month = {May},
      year = {2014},
      abstract = {Indoor localization is a hard problem for many
                physical reasons, and the majority of research
                attention has been devoted to devising mechanisms
                to achieve better spatial accuracy. However beyond
                gathering data, an equally important (but often
                ignored) question is what to do with location
                information when it is available. We propose
                semantic relations as building blocks for a first
                order language of localization. This
                representation may be used to express meaningful
                questions about the relative position of objects
                without necessarily resorting to physical
                coordinates. Since the same subset of a first
                order language can be used for different models
                when an elementary embedding exists between the
                backing structures, our approach does not restrict
                the actual representations of space provided they
                are so related. A heterogeneous strategy employing
                a combination of localization methods and
                different spatial representations for different
                areas can be accomplished by this method.},
      URL = {http://terraswarm.org/pubs/478.html}
  }
  

Posted by Matt Weber on 15 Jan 2015.
Groups: services