» Download Article

Authors
Luciene de Oliveira Marin; Mauro Roisenberg; Edson Roberto De Pieri

Identifier
10.3182/20060906-3-IT-2910.00013

Index Terms
reactive navigation,mobile robots,ART - adaptive resonance theory,multi-layer perceptron,reinforcement learning,unsupervised learning,on-line learning

Abstract
This paper describes a neural network architecture and the online learning policies that permits to an autonomous robot navigates though a maze in order to memorize a path that explores the entire environment, while avoiding obstacles. The state space representation is constructed by unsupervised and competitive learning as well as the mapping state-action is constructed by means of reinforcement learning, during the maze exploration. The result of learning creates a memory of states-actions that emerges an intelligent behavior, such as the path learning. The robot uses only its own infrared distance-sensors to perform obstacle detection, used as pattern recognition cues, while moving in a maze environment. In order to demonstrate the effectiveness and real-time ability of the proposed neural controller, we report a number of simulation results of navigation in unknown maze environments.

References

[1] Araújo, Rui and A. T. Almeida (1997). Mobile
    robot path-learning to separate goals on an
    unknown world. Intelligent Engineering Systems.
     INES '97. Proceedings. IEEE International
     Conference.
[2] Arleo, Angelo, Fabrizio Smeraldi and Wulfram
    Gerstner (2004). Cognitive navigation based
    on nonuniform gabor space sampling, unsupervised
     growing networks, and reinforcement
     learning. IEEE Transactions On Neural
    Networks 15(3), 639-652.
[3] Brignone, L. and M. Howarth (2003). Art-r: a
    novel reinforcement learning algorithm using
    an art module for state representation. Neural
     Networks for Signal Processing, 2003.
    NNSP'03. 2003 IEEE 13th Workshop pp. 829
    -838.
[4] Carpenter, Gail A. and Stephen Grossberg (1988).
    The ART of adaptive pattern recognition
    self-organizing by a neural network. IEEE
    Computer 21(3), 77-88.
[5] Cheng, Yu-Hu, Jian-Qiang Yi and Dongbin Zhao
    (2003). Sensor-based navigation system using
    a modified art-2 network and fuzzy controller
    for an indoor mobile robot. Proceedings of the
    Second International Conference on Machine
    Learning and Cybernetics, Xi'an pp. 1374-
    1379.
[6] Jan, Gene Eu, KI-Yin Chang and Ian Parberry
    (2003). A new maze routing approach for
    path planning of a mobile robot. Proceedings
    of the 2003 IEEE/ASME. International Conference
     on Advanced Intelligency Mechatronics
     (AIM 2003) pp. 552-557.
[7] Kaelbling, Leslie Pack, Michael L. Littman and
    Andrew W. Moore (1996). Reinforcement learning:
     A survey. Jounal of Artificial Intelligence
     Research pp. 237-285.
[8] Kohonen, Teuvo (1988). Self-Organization and
    Associative Memory. New York: Springer-Verlag.
    
[9] Lumelsky, Vladimir J. (1991). A comparative
    study on the path length performance of
    maze-searching and robot motion planning
    algorithms. IEEE Transactions on Robotics
    And Automation 7(1), 57-66.
[10] Martinengo, A., M. Campmi and V. Torre (1994).
     Artificial systems and complex behaviours.
     Intelligent Robots and Systems. Advanced Robotic
      Systems and the Real World, IROS'94.
     Proceedings of the IEEE/RSJ/GI International
      Conference on 1, 194-201.
[11] Mondada, F., E. Franzi and P. Lenne (1993). Mobile
      robot miniaturization: a tool for investigation
      in control algorithms. Int. Symposium
     on Experimental Robotics. Kyoto, Japan.
[12] Nehmzow, U. and B. McGonigle (1994). Achieving
      rapid adaptations in robots by means of
     external tuition. in: D.T. Cliff, P. Husbands,
     J.-A. Meyer, S. W. Wilson (Eds.), From Animals
      to Animats 3: Proceedings of the Adaptive
      Behavior, MIT Press, Cambridge, MA
     pp. 301-308.
[13] Noborio, Hiroshi, Keiichi Fhjimura and Yohei
     Horiuchi (2000). A comparative study of
     sensor-based path-planning algorithms in an
     unknown maze. Proceedings of the 2000
     IEEE, RSJ. International Conference on lntelligent
      Robots and Systems pp. 9009-9016.
[14] Omidvar, Omid and Patrick Van Der Smagt
     (1997). Neural Systems for Robotics. Academic
      Press.
[15] Pomerleau, Dean (1991). Efficient training of artificial
      neural networks for autonomous navigation.
      Neural Computation 3(1), 88-97.
[16] Rumelhart, D. E., G. E. Hinton and R. Williams
      (1986). Learning representations by back-propagation
      erros. Nature 323.
[17] Sehad, Samira and Claude Touzet (1995). Neural
      reinforcement path planning for the miniature
      robot khepera. Proceedings of the 1995
     World Congress on Neural Networks.
[18] Sutton, R. S. and A. G. Barto (1998). Reinforcement
      Learning: An Introduction. MIT Press.
[19] Zhu, Anmin and Simon X. Yang (2003). Self-organizing
      behavior of a multi-robot system
     by a neural network approach. Proceedings of
     the 2003 IEEE/RSJ Intl. Conference on Intelligent
      Robots and Systems pp. 1204-1209.