Journal of Intelligent and Robotic Systems

, Volume 15, Issue 2, pp 153–163 | Cite as

Self-organising neural networks for adaptive control

  • Kevin Warwick
  • Nigel Ball
Article

Abstract

Self-organizing neural networks have been implemented in a wide range of application areas such as speech processing, image processing, optimization and robotics. Recent variations to the basic model proposed by the authors enable it to order state space using a subset of the input vector and to apply a local adaptation procedure that does not rely on a predefined test duration limit. Both these variations have been incorporated into a new feature map architecture that forms an integral part of an Hybrid Learning System (HLS) based on a genetic-based classifier system. Problems are represented within HLS as objects characterized by environmental features. Objects controlled by the system have preset targets set against a subset of their features. The system's objective is to achieve these targets by evolving a behavioural repertoire that efficiently explores and exploits the problem environment. Feature maps encode two types of knowledge within HLS — long-term memory traces of useful regularities within the environment and the classifier performance data calibrated against an object's feature states and targets. Self-organization of these networks constitutes non-genetic-based (experience-driven) learning within HLS.

This paper presents a description of the HLS architecture and an analysis of the modified feature map implementing associative memory. Initial results are presented that demonstrate the behaviour of the system on a simple control task.

Key words

Neural networks adaptive control self-organising networks 

References

  1. 1.
    Ahalt, S., Krishnamurthy, A., Chen, P., and Mellon, D.: Competitive learning algorithms for vector Quantization, in Neural Networks, Vol. 3, Pergamon, Oxford, 1992, pp. 277–290.Google Scholar
  2. 2.
    Ball, N. and Warwick, K.: Using self-organizing feature maps for the control of artificial organisms, Proc. IEE, Part D. Vol. 140, No. 3, 1993.Google Scholar
  3. 3.
    Ball, N. and Warwick, K.: Application of augmented output self-organizing feature maps to the adaptive control problem, Proc. INNC-90 (Paris), No. 1, Kluwer Academic Publishers, Dordrecht, 1990.Google Scholar
  4. 4.
    Ball, N.: Organizing an animal's behavioral repertoires using Kohonen feature maps, in D. Cliff, P. Husbands, J. Meyer, and S. Wilson (eds), From Animals to Animals 3, MIT Press, 1994.Google Scholar
  5. 5.
    Cherkassky, V. and Hassein, L.: Self-organizing neural networks for non parametric regression analysis, Proc. INNC-90 (Paris), Vol. 1, Kluwer Academic Publishers, Dordrecht, 1990.Google Scholar
  6. 6.
    Holland, J. and Reitman, J.: Cognitive Systems based on adaptive algorithms, in D., Waterman and F., Hayes-Roth (eds), Pattern Directed Inference Systems, Academic Press, New York, 1978, pp. 313–329.Google Scholar
  7. 7.
    Holland, J., Holyoak, K., Nisbett, R., and Thagard, P.: Induction Processes of Inference, Learning and Discovery, MIT Press, 1986.Google Scholar
  8. 8.
    Kohonen, T.: Self Organization and Associative Memory, Springer-Verlag, 1984.Google Scholar
  9. 9.
    Whittington, G. and Spraken, T.: The application of neural networks to industrial spectral analysis, identification and classification, Proc. IEEE Workshop on Genetic Algorithms, Simulated Annealing and Neural Nets., University of Glasgow, 1990.Google Scholar

Copyright information

© Kluwer Academic Publishers 1996

Authors and Affiliations

  • Kevin Warwick
    • 1
  • Nigel Ball
    • 2
  1. 1.Department of Cybernetics, School of Engineering and Information SciencesUniversity of ReadingReadingUK
  2. 2.Engineering Design Centre, Department of EngineeringUniversity of CambridgeCambridgeUK

Personalised recommendations