Comparing Cellular and Panmictic Genetic Algorithms for Real-Time Object Detection

  • Jesús Martínez-Gómez
  • José Antonio Gámez
  • Ismael García-Varea
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6024)


Object detection is a key point in robotics, both in localization and robot decision making. Genetic Algorithms (GAs) have proven to work well in this type of tasks, but they usually give rise to heavy computational processes. The scope of this study is the Standard Platform category of the RoboCup soccer competition, and so real-time object detection is needed. Because of this, we constraint ourselves to the use of tiny GAs. The main problem with this type of GAs is their premature convergence to local optima. In this paper we study two different approaches to overcoming this problem: the use of population re-starts, and the use of a cellular GA instead of the standard generational one. The combination of these approaches with a clever initialisation of the population has been analyzed experimentally, and from the results we can conclude that for our problem the best choice is the use of cellular GAs.


Genetic Algorithm Local Optimum Object Detection Memetic Algorithm Biped Robot 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Alba, E., Dorronsoro, B.: Cellular genetic algorithms, vol. 1 (March 2008)Google Scholar
  2. 2.
    Bach, J., Jungel, M.: Using pattern matching on a flexible, horizon-aligned grid for robotic vision. Concurrency, Specification and Programming-CSP 1(2002), 11–19 (2002)Google Scholar
  3. 3.
    Coath, G., Musumeci, P.: Adaptive arc fitting for ball detection in robocup. In: APRS Workshop on Digital Image Analysing. pp. 63–68 (2003)Google Scholar
  4. 4.
    Fukunaga, A.: Restart scheduling for genetic algorithms. In: Eiben, A.E., Bäck, T., Schoenauer, M., Schwefel, H.-P. (eds.) PPSN 1998. LNCS, vol. 1498, pp. 357–366. Springer, Heidelberg (1998)CrossRefGoogle Scholar
  5. 5.
    Jüngel, M., Hoffmann, J., Lötzsch, M.: A real-time auto-adjusting vision system for robotic soccer. In: Polani, D., Browning, B., Bonarini, A., Yoshida, K. (eds.) RoboCup 2003. LNCS (LNAI), vol. 3020, pp. 214–225. Springer, Heidelberg (2004)Google Scholar
  6. 6.
    Martínez-Gómez, J., Gámez, J.A., García-Varea, I., Matellán, V.: Using genetic algorithm for real-time object detection. In: Proceedings of the 2009 RoboCup Symposium, pp. 215–227. Springer, Heidelberg (2009)Google Scholar
  7. 7.
    Mitchell, M.: An Introduction to Genetic Algorithms. MIT Press, Cambridge (1998)zbMATHGoogle Scholar
  8. 8.
    Moscato, P.: Memetic algorithms: a short introduction. Mcgraw-Hill’S Advanced Topics in Computer Science Series, pp. 219–234 (1999)Google Scholar
  9. 9.
    Rofer, T., Brunn, R., Dahm, I., Hebbel, M., Hoffmann, J., Jungel, M., Laue, T., Lotzsch, M., Nistico, W., Spranger, M.: GermanTeam 2004. Team Report RoboCup (2004)Google Scholar
  10. 10.
    Wasik, Z., Saffiotti, A.: Robust color segmentation for the robocup domain. In: Proc. of the Int. Conf. on Pattern Recognition (ICPR), vol. 2, pp. 651–654 (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Jesús Martínez-Gómez
    • 1
  • José Antonio Gámez
    • 1
  • Ismael García-Varea
    • 1
  1. 1.Computing Systems Department, SIMD i3AUniversity of Castilla-la ManchaAlbaceteSpain

Personalised recommendations