Advertisement

On Interpolating Memories for Learning Control

  • H. Tolle
  • S. Gehlen
  • M. Schmitt
Chapter
Part of the Advances in Industrial Control book series (AIC)

Abstract

To imitate human flexibility in controlling different complex non-linear processes on the basis of process observation and/or trial and error, learning control has been developed. The main elements of such control loops are interpolating memories. The chapter deals after an introduction to learning control loops with such devices by putting forward different alternatives, discussing their behaviour in general and going into details of recent research work on mathematically inspired interpolating memories. The respective improvements are motivated and results of applications in the areas of biotechnology and automotive control are presented. In a conclusion some further application areas and realisation aspects are discussed and a critical assessment of status and usefulness of learning control is given.

Keywords

Input Space Regular Grid Scattered Data Feedforward Control Training Point 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    J.S. Albus. Theoretical and Experimental Aspects of a Cerebellar Model. PhD thesis, University of Maryland, Maryland, 1972.Google Scholar
  2. 2.
    J.S. Albus. A new approach to manipulator control: The cerebellar model articulation controller. Transactions ASME, 97(3), 1975.Google Scholar
  3. 3.
    Andrew R. Barron. Universal approximation bounds for superpositions of a sigmoidal function. IEEE Transactions on Information Theory, 39(3):930–945, May 1993.zbMATHCrossRefGoogle Scholar
  4. 4.
    Martin Brown and Christopher J. Harris. Neurofuzzy Adaptive Modelling and Control. Prentice Hall, ISBN 0–13–134453–6, 1994.Google Scholar
  5. 5.
    E. Ersü and J. Militzer. Software implementation of a neuron-like associative memory system for control applications. In 2nd IASTED Conference on Mini- and Micro-Computer Applications — MIMI’82. Davos, Switzerland, March 1982.Google Scholar
  6. 6.
    Enis Ersü and Jürgen Militzer. Real-time implementation of an associative memory-based learning control scheme for non-linear multivariable processes. In IEE-Symposium: Application of Multivariable System Techniques, Plymouth, UK, 31. Oktober – 2. November 1984.Google Scholar
  7. 7.
    P. Funk. Variationsrechnung und ihre Anwendung in Physik und Technik. Springer Verlag, 2 edition, 1970.zbMATHGoogle Scholar
  8. 8.
    Stefan Gehlen. Untersuchungen zur wissensbasierten und lernenden Prozeßführung in der Biotechnologie. PhD thesis, TH Darmstadt, FG Regelsystemtheorie & Robotik, 1993. Fortschritt-Berichte VDI, Reihe 20, Rechnerunterstützte Verfahren, Nr. 87, VDI-Verlag, ISBN 3–18–148720–1.Google Scholar
  9. 9.
    C. J. Harris, C. G. Moore, and M. Brown. Intelligent Control — Aspects of fuzzy logic and neural nets. World scientific, 1993.zbMATHGoogle Scholar
  10. 10.
    Rolf Isermann. Digital Control Systems. Springer, 1981.zbMATHGoogle Scholar
  11. 11.
    A. G. Ivankhenko. Heuristic self-organization in problems of engin. cybernetics. Automatica, 6, 1970.Google Scholar
  12. 12.
    K. Kleinmann, M. Hormel, and W. Paetsch. Intelligent real-time control of a multifingered robot gripper by learning incremental actions. In IFAC/IFIP/IMACS Int. Symp. on Artificial Intelligence in Real-Time Control. Delft, The Netherlands, June 1992.Google Scholar
  13. 13.
    K. Kleinmann and R. Wacker. On a selftuning decoupling controller for the joint control of a tendon driven multifingered robot gripper. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS’94). Munich, 1994.Google Scholar
  14. 14.
    M. Kortmann and H. Unbehauen. Ein neuer Algorithmus zur automatischen Selektion der optimalen Modellstruktur bei der Identifikation nichtlinearer Systeme. Automatisierungstechnik (at), 1987.Google Scholar
  15. 15.
    A. Kurz. Building maps based on a learned classification of ultrasonic range data. In D. Charnley, editor, 1st IFAC Workshop on Intelligent Autonomous Vehicles. Pergamon Press, Southampton, Southampton, UK, April 1993.Google Scholar
  16. 16.
    R. P. Lippmann. An introduction to computing with neural nets. IEEE ASSP Magazine, April 1987.Google Scholar
  17. 17.
    Jürgen Militzer and Henning Tolle. Vertiefungen zu einem Teilbereiche der menschlichen Intelligenz imitierenden Regelungsansatz. In Jahrestagung der Deutschen Gesellschaft für Luft- und Raumfahrt, München, 1986.Google Scholar
  18. 18.
    W. Thomas Miller III, Filson H. Glanz, and L. Gordon Kraft III. Application of a general learning algorithm to the control of robotic manipulators. The International Journal of Robotics and Control, 6(2):84–98, 1987.CrossRefGoogle Scholar
  19. 19.
    W. S. Mischo, M. Hormel, and H. Tolle. Neurally inspired associative memories for learning control. A comparison. In ICANN — 91, International Conference on Artificial Neural Networks. Espoo, Finland, June 1991.Google Scholar
  20. 20.
    Walter Sebastian Mischo and Henning Tolle. Ein assoziativer VLSI-Prozessor zur schnellen Informations-/Stellsignalgenerierung. In Fachtagung Integrierte mechanisch-elektronische Systeme, number 179 in VDI Fortschrittsberichte, pages 263–278. VDI Verlag, 1993.Google Scholar
  21. 21.
    W. Paetsch and M. Kaneko. A three fingered, multijoined gripper for experimental use. In IROS’90, Int. Workshop on Intelligent Robots and Systems. Tsuchiusa, Ibaraki, Japan, July 1990.Google Scholar
  22. 22.
    Jürgen Roth, Bert Breuer, and Jörg Stöcker. Kraftschlußerkennung im rotierenden Reifen. In Fachtagung Integrierte mechanisch-elektronische Systeme, number 179 in VDI Fortschrittsberichte, pages 132–143. VDI Verlag, 1993.Google Scholar
  23. 23.
    G. N. Saridis. Self-Organizing Control of Stochastic Systems. M. Dekker, 1977.zbMATHGoogle Scholar
  24. 24.
    M. Schmitt and H. Tolle. Das Assoziativkennfeld, eine lernfähige Standardkomponente für Kfz-Steuergeräte. ATZ (Automobiltechnische Zeitschrift), 94(1), 1994.Google Scholar
  25. 25.
    Manfred Schmitt. Untersuchungen zur Realisierung mehrdimensionaler, lernfähiger Kennfelder in Großserien-Steuergeräten. PhD thesis, TH Darmstadt, 1994. — in preparation -.Google Scholar
  26. 26.
    H. Tolle, P. C. Parks, E. Ersü, M. Hormel, and J. Militzer. Learning control with interpolating memories — general ideas, design-lay-out, theoretical approaches and practical applications. Int. J. Control, 56, 1992.Google Scholar
  27. 27.
    Henning Tolle and Enis Ersü. Neurocontrol Number 172 in Lecture Notes in Control and Information Sciences. Springer-Verlag, 1992. ISBN 3–540–55057–7.Google Scholar

Copyright information

© Springer-Verlag London Limited 1995

Authors and Affiliations

  • H. Tolle
    • 1
  • S. Gehlen
    • 2
  • M. Schmitt
    • 3
  1. 1.Institute of Control EngineeringTechnical University DarmstadtGermany
  2. 2.Zentrum für Neuroinformatik GmbHBochumGermany
  3. 3.R. Bosch GmbHDept. Motor Vehicle Safety SystemsSchwieberdingenGermany

Personalised recommendations