Advertisement

Understanding Control Strategies

  • Ivan Bratko
  • Dorian Šuc
Part of the International Centre for Mechanical Sciences book series (CISM, volume 472)

Abstract

Controlling complex dynamic systems requires skills that operators often cannot completely describe, but can demonstrate. This paper describes research into the understanding of such tacit control skills. Understanding tacit skills has practical motivation in respect of communicating skill to other operators, operator training, and also mechanising and optimising human skill. This paper is concerned with approaches to the understanding of human operators’ skill by analysing operators’ traces using techniques of machine learning (ML). The paper gives a review of ML-based approaches to skill reconstruction. Recent work is presented with particular emphasis on understanding human tacit skill qualitatively, and automatically generating explanation of how it works. This includes qualitative machine learning to extract from operator’s control traces his subconscious sub-goals and control strategies described in qualitative terms.

Keywords

Crane System Gantry Crane Automatic Controller Control Skill Qualitative Simulation 
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. Bain, M., Sammut, C. (1999) In: Machine Intelligence 15, (eds. K. Furukawa, D. Michie, S. Muggleton ), Oxford University Press.Google Scholar
  2. Bratko, I. (1997) Qualitative reconstruction of control skill. Proc. QR97 (11th Int. Workshop on Qualitative Reasoning), pp. 41–52. Cortona, Italy, 1997 (Pubblicazioni N. 1036. Pavia: Instituto di analisi numerica.Google Scholar
  3. Bratko, I. (2001) Prolog Programming for Artificial Intelligence, 3rd edition. Addison-Wesley.Google Scholar
  4. I. Bratko, T. Urbaneic (1999) Control skill, machine learning and hand-crafting in controller design. In: Machine Intelligence 15, (eds. K. Furukawa, D. Michie, S.Muggleton ), Oxford University Press.Google Scholar
  5. I. Bratko, T. Urbana, C. Sammut (1998) Behavioural cloning of control skill In: Machine Learning and Data Mining: Methods and Applications, (eds. R.S. Michalski, I. Bratko, M. Kubat) Wiley.Google Scholar
  6. Breiman, L., Friedman, J H., Olshen, R. A., Stoe, C. J. (1984) Classification and Regression Trees. Belmont, CA: Wadswarth.MATHGoogle Scholar
  7. Karalic, A. (1992) Employing linear regression in regression tree leaves. Proc. 10th European Conf. on Artificial Intelligence, pp. 440–441. Vienna, 1992.Google Scholar
  8. Karaliò, A., Bratko, I. (1997) First Order Regression. Machine Learning, Vol. 26, 147–176.CrossRefGoogle Scholar
  9. Krizman, V., Dzeroski, S. and Kompare B. (1995) Discovering dynamics from measured data. Electrotechnical Review, Vol. 62, 19–198.Google Scholar
  10. Kuipers, B. (1994) Qualitative Reasoning: Modeling and Simulation with Incomplete Knowledge. Cambridge, MA: MIT Press.Google Scholar
  11. Michie, D. (1993) Knowledge, learning and machine intelligence. In: L.S.Sterling, (ed.) Intelligent Systems, Plenum Press.Google Scholar
  12. Michie, D., Bain, M., Hayes-Michie, J. (1990) Cognitive models from subcognitive skills. In: Grimble, M., McGhee, J., Mowforth, P. (eds.), Knowledge-Based Systems in Industrial Control, Stevenage: Peter Peregrinus.Google Scholar
  13. Michie, D., Camacho, R. (1994) Building symbolic representations of intuitive real-time skills from performance data. In: K. Furukawa, S. Muggleton (eds.), Machine Intelligence and Inductive Learning, Oxford: Oxford University Press.Google Scholar
  14. Quinlan, J. R. (1986) Induction of decision trees. Machine Learning 1: 81–106.Google Scholar
  15. Sammut, C. and Hurst, S. and Kedzier, D. and Michie, D. (1992) Learning to fly. Proc. IC AIL ‘82 (9th Int. Conf. on Machine Learning), Aberdeen, 1992.Google Scholar
  16. Šuc, D. (2001) Machine Reconstruction of Human Control Strategies. Ph. D. Thesis, Univ. of Ljubljana, Faculty of Computer and Information Sc.Google Scholar
  17. Šuc, D., Bratko, I. (1997) Reconstructing control skill as LQ controllers with subgoals Proc. IJCAI’97, Yokohama, Japan, August 1997.Google Scholar
  18. Šuc, D., Bratko, I. (1999) Symbolic and qualitative reconstruction of control skill. ETAI Journal (Electronic Transactions on Artificial Intelligence), Vol. 3 (1999), Section B, pp. 1–22. http://www.ep.liu.se/ej/etai/1999/002/.
  19. Šuc, D., Bratko, I. (2000a) Problem decomposition for behavioural cloning Proc. ECML’2000 (Europena Conf. Machine Learning), Barcelona, June 2000Google Scholar
  20. Šuc, D., Bratko, I. (2000b) Skill modeling through symbolic reconstruction of operator’s trajectories. IEEE trans. syst. man cybern., Part A, Syst. humans, 2000, vol$130, no. 6, str. 617–624.Google Scholar
  21. Šuc, D., Bratko, I. (2000c) Qualitative trees applied to bicycle riding. ETAI Journal (Electronic Transactions on Artificial Intelligence), Vol. 4 (2000), Section B, pp. 125–140. http://www.ep.liu.se/ej/etai/2000/014/.
  22. Šuc, D., Bratko, I. (2001) Induction of qualitative trees. ECML’2000 (Europena Conf. Machine Learning), Barcelona, June 2000Google Scholar
  23. Urbancic, T., Bratko, I. (1994) Reconstructing human subcognitive skill through machine learning. Proc. ECA1–94, pp. 498–502. Amsterdam, 1994.Google Scholar

Copyright information

© Springer-Verlag Wien 2003

Authors and Affiliations

  • Ivan Bratko
    • 1
  • Dorian Šuc
    • 1
  1. 1.Faculty of Computer and Information Sc.University of LjubljanaLjubljanaSlovenia

Personalised recommendations