Methods for evolving robust distributed robot control software: coevolutionary and single population techniques

Created by W.Langdon from gp-bibliography.bib Revision:1.3973

@InProceedings{dolin:2001:eh,
  author =       "Brad Dolin and Forrest H {Bennett III} and 
                 Eleanor G. Rieffel",
  title =        "Methods for evolving robust distributed robot control
                 software: coevolutionary and single population
                 techniques",
  booktitle =    "The Third NASA/DoD workshop on Evolvable Hardware",
  year =         "2001",
  editor =       "Didier Keymeulen and Adrian Stoica and Jason Lohn and 
                 Ricardo S. Zebulum",
  pages =        "21--29",
  address =      "Long Beach, California",
  publisher_address = "1730 Massachusetts Avenue, N.W., Washington, DC,
                 20036-1992, USA",
  month =        "12-14 " # jul,
  organisation = "Jet Propulsion Laboratory, California Institute of
                 Technology",
  publisher =    "IEEE Computer Society",
  keywords =     "genetic algorithms, genetic programming,
                 coevolutionary approaches, coevolutionary population
                 techniques, distributed control software, modular
                 robot, modular robots, random sampling, robust
                 distributed robot control software, single population
                 techniques, control engineering computing, distributed
                 control, robots",
  ISBN =         "0-7695-1180-5",
  DOI =          "doi:10.1109/EH.2001.937943",
  size =         "9 pages",
  abstract =     "Previous work on evolving distributed control software
                 for modular robots has resulted in solutions that do
                 not generalise well to unseen test cases. In this work,
                 we seek general solutions to an entire space of test
                 cases. Each test case is a specific world configuration
                 with a passage through which the modular robot must
                 move. The space of test cases is extremely large, so a
                 given training set can only be a sparse sample of this
                 space. We look at several approaches for dealing with
                 the problem of determining an effective training set:
                 using a fixed set throughout a run, sampling randomly
                 at each generation, and using coevolutionary approaches
                 to evolve a population of test worlds. For this
                 problem, random sampling outperformed the fixed
                 sampling technique and did at least as well as the
                 coevolutionary techniques we considered",
  notes =        "EH2001 http://cism.jpl.nasa.gov/ehw/events/nasaeh01/
                 Note misspeling of Brad Dolin as {"}Dofin, B.{"}.",
}

Genetic Programming entries for Brad Dolin Forrest Bennett Eleanor G Rieffel

Citations