Advertisement

Knowledge extraction using neural network by an artificial life approach

  • Yuji Makita
  • Masafurni Hagiwara
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1285)

Abstract

A novel knowledge extraction method from autonomous behavior of multiple mobile robots by an artificial life approach is proposed in this paper. The knowledge is expressed by if-then rules and we employ a neural network for the knowledge extraction. The proposed method has the following features: 1)The structure of knowledge extraction neural network and the learning algorithm are simple; 2)Understanding and modification of the extracted knowledge are easy because weights in the knowledge extraction network directly represent the antecedents and the consequents of the if-then rules; 3)The network itself has an ability of inference using the extracted knowledge. We used a lot of autonomous mobile robots in various environments. Each robot has to avoid the obstacles to get to the goal and the local behavior is extracted and integrated in the knowledge extraction neural network as global knowledge. We confirmed the validity of the proposed method by computer simulations.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    P.H.Winston (ed.), Artificial Intelligence (3rd edition), Addison-Wesley, 1988.Google Scholar
  2. 2.
    J.L.McClelland, D.E.Rumelhart and the PDP Research Group, Parallel Distributed Processing, The MIT Press, 1986.Google Scholar
  3. 3.
    B.Kosko, Neural Networks and Fuzzy Systems, Prentice Hall, 1992.Google Scholar
  4. 4.
    D.E.Goldberg, Genetic Algorithms in Search, Optimization, and Machine Learning, Addison-Wesley, 1989.Google Scholar
  5. 5.
    A.Lindenmayer and P.Prusinkiewicz, “Developmental models of multicellular organisms: a computer graphics perspective,” in C.G.Langton et al.(eds.), Artificial Life II, Addison-Wesley, 1992.Google Scholar
  6. 6.
    T.S.Ray, “An approach to the synthesis of life,” in C.G.Langton et al.(eds.), Artificial Life II, Addison-Wesley, 1992.Google Scholar
  7. 7.
    R.A.Brooks, “A robust layered control system for a mobile robot,” IEEE J. Robotics and Automation, vol. 2, pp. 14–23, 1986.Google Scholar
  8. 8.
    T.Shibata and T.Fukuda, “Hierarchical intelligent control for robotic motion,” IEEE Trans. Neural Networks, vol. 5, pp. 823–832, 1994.Google Scholar
  9. 9.
    M.Dorigo, V.Maniezzo and A.Colorni, “Ant system: optimization by a colony of cooperating agents,” IEEE Trans. Systems, Man and Cybernetics, vol. 26-B, pp. 29–41, 1996.Google Scholar
  10. 10.
    D.Ackley and M.Littman, “Interactions between learning and evolution,” in C.G.Langton et al.(eds.), Artificial Life II, Addison-Wesley, 1992.Google Scholar
  11. 11.
    J.Koza, “Genetic programming, on the programming of computers by means of natural selection,” MIT Press, 1992.Google Scholar
  12. 12.
    S.-G. Kong and B.Kosko, “Adaptive fuzzy systems for backing up a truck-andtrailer,” IEEE Trans. Neural Networks, vol. 3, pp. 211–223, Mar. 1992.Google Scholar
  13. 13.
    S.Tadokoro, M.Hayashi, Y.Manabe, Y.Nakami and T.Takamori, “Motion planner of mobile robots which avoid moving human obstacles on the basis of stochastic prediction,” IEEE Int. Conf. Systems, Man and Cybernetics, vol. 4, pp. 3286–3291, 1995.Google Scholar
  14. 14.
    J.A.Adams and R.Paul, “Human supervisory control of multiple mobile agents,” IEEE Int. Conf. Systems, Man and Cybernetics, vol. 4, pp. 3298–3303, 1995.Google Scholar
  15. 15.
    T.Yamaguchi, S.Sekine, D.Montgomery and T.Endo, “Intelligent interface based on fuzzy associative inference and its application to command spelling corrector,” T.IEE Japan, vol. 113-C, pp. 709–718, 1993.Google Scholar
  16. 16.
    E.S.Plumer, “Neural network structure for navigation using potential fields,” in Proc. Int. Joint Conf. Neural Networks (IJCNN-9,2), vol. 1, pp. 327–332, 1992.Google Scholar
  17. 17.
    K.Sato, “Global motion planning using a laplacian potential field,” JRSJ, vol. 11, pp. 702–709, 1993.Google Scholar
  18. 18.
    Y.Makita and M.Hagiwara, “A path planning system using fuzzy rules and a potential field,” T.IEE Japan, vol. 115-C, pp. 772–778, 1995.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Yuji Makita
    • 1
  • Masafurni Hagiwara
    • 1
  1. 1.Department of Electrical EngineeringKeio UniversityKohoku-ku, YokohamaJapan

Personalised recommendations