Open-ended evolution to discover analogue circuits for beyond conventional applications

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

@Article{Sapargaliyev:2012:GPEM,
  author =       "Yerbol A. Sapargaliyev and Tatiana G. Kalganova",
  title =        "Open-ended evolution to discover analogue circuits for
                 beyond conventional applications",
  journal =      "Genetic Programming and Evolvable Machines",
  year =         "2012",
  volume =       "13",
  number =       "4",
  pages =        "411--443",
  month =        dec,
  keywords =     "genetic algorithms, genetic programming, evolvable
                 hardware, Analogue Circuit Synthesis, CAD, SPICE,
                 Simulation, Evolutionary algorithms",
  ISSN =         "1389-2576",
  DOI =          "doi:10.1007/s10710-012-9163-8",
  URL =          "http://results.ref.ac.uk/Submissions/Output/259508",
  size =         "33 pages",
  abstract =     "Analogue circuits synthesised by means of open-ended
                 evolutionary algorithms often have unconventional
                 designs. However, these circuits are typically highly
                 compact, and the general nature of the evolutionary
                 search methodology allows such designs to be used in
                 many applications. Previous work on the evolutionary
                 design of analogue circuits has focused on circuits
                 that lie well within analogue application domain. In
                 contrast, our paper considers the evolution of analogue
                 circuits that are usually synthesised in digital logic.
                 We have developed four computational circuits, two
                 voltage distributor circuits and a time interval metre
                 circuit. The approach, despite its simplicity, succeeds
                 over the design tasks owing to the employment of
                 substructure reuse and incremental evolution. Our
                 findings expand the range of applications that are
                 considered suitable for evolutionary electronics.",
  notes =        "PhD in ES EHW 2011",
  affiliation =  "School of Engineering and Design, Brunel University,
                 Uxbridge, Middlesex UB83PH, UK",
  uk_research_excellence_2014 = "This paper presents a novel solution
                 method combining classical genetic algorithm,
                 substructure reuse and incremental evolution to a
                 well-known problem in EHW community. The method was
                 implemented using C++ and PSpice, as requested by
                 Creative Machines Lab, Cornell University, USA. The
                 approach is based on simulation of analogue circuit for
                 a given graphically described signal. If the problem
                 can be slightly rephrased as the analysis of sound
                 waves for pitch identification, this approach can be
                 successfully applied to completely new problem domains.
                 This work led to an on-going three-year project on
                 Sound Recognition fully sponsored by
                 Scientific-Technical Centre REB, Kazakhstan.",
}

Genetic Programming entries for Yerbol A Sapargaliyev Tatiana Kalganova

Citations