Novelty Search for software improvement of a SLAM system

Created by W.Langdon from gp-bibliography.bib Revision:1.4504

  author =       "Victor R. Lopez-Lopez and Leonardo Trujillo and 
                 Pierrick Legrand",
  title =        "Novelty Search for software improvement of a {SLAM}
  booktitle =    "5th edition of GI @ GECCO 2018",
  year =         "2018",
  editor =       "Brad Alexander and Saemundur O. Haraldsson and 
                 Markus Wagner and John R. Woodward and Shin Yoo",
  pages =        "1598--1605",
  address =      "Kyoto, Japan",
  month =        "15-19 " # jul,
  organisation = "ACM SIGEvo",
  publisher =    "ACM",
  keywords =     "genetic algorithms, genetic programming, genetic
                 improvement, vision for robotics, SLAM, Novelty
  URL =          "",
  DOI =          "doi:10.1145/3205651.3208237",
  size =         "8 pages",
  abstract =     "Genetic Improvement (GI) performs a search at the
                 level of source code to find the best variant of a
                 baseline system that improves non-functional properties
                 while maintaining functionality, with noticeable
                 results in several domains. There a many aspects of
                 this general approach that are currently being
                 explored. In particular, this work deals to the way in
                 which the search is guided to efficiently explore the
                 search space of possible software versions in which GI
                 operates. The proposal is to integrate Novelty Search
                 (NS) within the GISMOE GI framework to improve
                 KinectFusion, which is a vision-based Simultaneous
                 Localization and Mapping (SLAM) system that is used for
                 augmented reality, autonomous vehicle navigation, and
                 many other real-world applications. This is one of a
                 small set of works that have successfully combined NS
                 with a GP system, and the first time that it has been
                 used for software improvement. To achieve this, we
                 propose a new behaviour descriptor for SLAM algorithms,
                 based on state-of-the-art benchmarking and present
                 results that show that NS can produce significant
                 improvement gains in a GI setting, when considering
                 execution time and trajectory estimation as the main
                 performance criteria.",
  notes =        "SLAMbench, GPU KinectFusion, CUDA. C++, GISMOE, BNF
                 grammar, 200 generations. KVM, Ubuntu 16.0, GNU
                 Parallel. ICL-NUM videos 2 and 4 are used to train.
                 'improvements on both ATE and EXT, of 26.3percent and


Genetic Programming entries for Victor Raul Lopez Lopez Leonardo Trujillo Pierrick Legrand