Advertisement

Evolutionary Procedure for the Progressive Design of Controllers for Collective Behaviors

  • P. Caamaño
  • J. A. Becerra
  • F. Bellas
  • A. Prieto
  • R. J. Duro
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6679)

Abstract

This paper describes an approach for the progressive construction of controllers for sets of robots performing collective behaviors. The procedure is based on the incremental construction through evolution of a neural multilevel behavior architecture where the higher-level behaviors modulate the actuation of the lower-level ones. This hybridization permits simplifying the design of the behavior controllers and allows obtaining them in evolutionary processes without making the search space huge. From a cognitive point of view, the procedure could be thought of as an incremental learning procedure where the robot first learns basic responses and then uses them within more elaborate decision and actuation processes progressively increasing complexity.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Kolling, A., Carpin, S.: Multi-robot Surveillance: an Improved Algorithm for the GRAPH-CLEAR Problem. In: Proc. 2008 IEEE ICRA Pasadena, CA, USA, pp. 2360–2365 (2008)Google Scholar
  2. 2.
    Fernandez-Gauna, B., Lopez-Guede, J.M., Zulueta, E., Graña, M.: Learning hose transport control with Q-learning. Neural Network World 20(7), 913–923 (2010)Google Scholar
  3. 3.
    Burgard, W., Moors, M., Stachniss, C., Schneider, F.: Coordinated multi-robot exploration Robotics. IEEE Transactions on Robotics 21(3), 376–386 (2005)CrossRefGoogle Scholar
  4. 4.
    Folgado, E., Rincón, M., Álvarez, J.R., Mira, J.: A Multi-robot Surveillance System Simulated in Gazebo. In: Mira, J., Álvarez, J.R. (eds.) IWINAC 2007. LNCS, vol. 4528, pp. 202–211. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  5. 5.
    Waibel, M., Keller, L., Floreano, D.: Genetic Team Composition and Level of Selection in the Evolution of Multi-Agent Systems. IEEE TEC 13(3), 648–660 (2009)Google Scholar
  6. 6.
    José Antonio Martín, H., de Lope, J., Maravall, D.: Learning to coordinate multi-robot competitive systems by stimuli adaptation. In: Mira, J., Ferrández, J.M., Álvarez, J.R., de la Paz, F., Toledo, F.J. (eds.) IWINAC 2009. LNCS, vol. 5602, pp. 362–371. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  7. 7.
    Cavalcanti, A., Freitas Jr., R., Kretly, L.: Nanorobotics Control Design: A Practical Approach Tutorial, Robotics Today, SME, 4th Quarter, vol.18(4) (2005)Google Scholar
  8. 8.
    Graña, M., Torrealdea, F.J.: Hierarchically structured Systems. European Journal of Operational Research 25, 20–26 (1986)CrossRefGoogle Scholar
  9. 9.
    Becerra, J.A., Bellas, F., Santos, J., Duro, R.J.: Complex Behaviours through modulation in Autonomous Robot Control. In: Cabestany, J., Prieto, A.G., Sandoval, F. (eds.) IWANN 2005. LNCS, vol. 3512, pp. 717–723. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  10. 10.
    Duro, R.J., Santos, J.: Discrete Time Backpropagation for training Synaptic Delay Based Artificial Neural Networks. IEEE Trans. on Neural Networks 10, 779–789 (1999)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • P. Caamaño
    • 1
  • J. A. Becerra
    • 1
  • F. Bellas
    • 1
  • A. Prieto
    • 1
  • R. J. Duro
    • 1
  1. 1.Integrated Group for Engineering ResearchUniversidade da CoruñaSpain

Personalised recommendations