• DocumentCode
    3723226
  • Title

    Evolutionary Robotics on Lego NXT Platform

  • Author

    Poikselk?;Ilari Vallivaara; R?ning

  • Author_Institution
    Dept. of Comput. Sci. &
  • fYear
    2015
  • Firstpage
    1137
  • Lastpage
    1144
  • Abstract
    This paper studies the Lego NXT platform´s suitability for evolutionary robotics. It is shown that the low-cost Lego NXT educational set is indeed adequate for simple experiments in evolutionary robotics. This is demonstrated by an experiment, where an artificial neural network-based controller capable of behaving meaningfully in a Lego sumo wrestling context is evolved on physical Lego NXT robots without the aid of simulation. A detailed description of the experiment is provided, and the practical aspects of actually conducting evolutionary robotics on the platform are studied. Earlier research suggests that using evolutionary robotics in education could provide a good and concrete example of the principles and mechanisms of evolution. The research described here utilizes only standard Lego NXT Educational kits, making conducting the experiments possible for a very wide audience. To the authors´ knowledge this is the first time the non-simulated Lego NXT is used to conduct artificial neural network-based evolutionary robotics.
  • Keywords
    "Robot sensing systems","Neurons","Acoustics","Mobile robots","Biological neural networks"
  • Publisher
    ieee
  • Conference_Titel
    Tools with Artificial Intelligence (ICTAI), 2015 IEEE 27th International Conference on
  • ISSN
    1082-3409
  • Type

    conf

  • DOI
    10.1109/ICTAI.2015.161
  • Filename
    7372258