Abstract
The BOXES paradigm has been successfully applied to control systems that are mechanically unstable. The primary application is the “Trolley and Pole” in which a freely hinged pole is balanced by rapidly reversing the direction of a guided trolley. This paper takes the BOXES (Michie [1]) methodology into the realm of continuous control systems, in which success is not measured by time to failure, but rather by the form factors of response to stimuli. To illustrate the method a classic second order, damped harmonic system is modified to include an AI contribution to its control parameters. The paper establishes that such contributions are both significant and desirable. The paper concludes with some encouragements for further study of systems that are poorly or partially defined.
Keywords
- Automatic Control System
- Control Matrix
- Conventional Control
- Loop Transfer Function
- Generation Artificial Intelligence
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.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Bibliography
Michie, D. & Chambers, R.A. Machine Intelligence 2…BOXES: An experiment in Adaptive Control. Oliver & Boyd
Russell, D.W. The Trolley & Pole Revisited: Further Studies in AI Control of a Mechanically Unstable System. AIENG’90. Boston, May 1990.
Russell, D.W. & Rees, S. J. System Control: A Case Study of a Statistical Learning Automaton. Proc. 2nd European Meeting on Cybernetics & Systems Research. Vienna, 1974.
Russell, D.W., Rees, S.J. & Boyes, J.A. Microsystem for Control by Automata of Real Life Situations. Proc. C.I.S.S. Baltimore, 1977.
Kuo, Benjamin C. Automatic Control Systems 6 ed Chapter 8.3. Time Domain Design of Control Systems, pp. 480 ff. Prentice Hall. Englewood Cliffs, N.J. 1991.
Bain, M. Michie D., & Sammut, C. Experiments with the Pole Balancing System: A Suitable Case for Genetic Treatment. Proc. ISSEK, Udine, Italy, 1988.
Boden, M.A.. Failure is not the Spur Adaptive Control of Defined Systems. Proc. NATO Adv. Research Institute, June 1981, Moretonhampstead, Devon. Plenum Press, N.Y. 1984.
Rees, S.J. An Investigation Into Possible Applications of Learning Control Systems. M. Phil thesis. Liverpool Polytechnic, 1978.
Boyes. J. A Man-machine Interface for Training an Automation for a Real World Situation. M. Phil thesis. Liverpool Polytechnic, 1977.
DiAngelo, J. Trash to Steam Optimization Using the BOXES Methodology, Unpublished Masters Paper. Penn State Great Valley, 1991.
Brownlowe, W. Control Strategies for Microbial Fermentation, Unpublished Master’s Paper. Penn State Great Valley, 1991.
Barto, A.G., Sutton R.S. & Anderson C.W. Neuronlike Adaptive Elements That Can Solve Difficult Learning Control Problems. IEEE Transactions on Systems, Man & Cybernetics. SMC — 13: 834–846. 1983.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1991 Computational Mechanics Publications
About this chapter
Cite this chapter
Russell, D.W. (1991). Studies in A.I. Augmented Control Systems using the Boxes Methodology. In: Rzevski, G., Adey, R.A. (eds) Applications of Artificial Intelligence in Engineering VI. Springer, Dordrecht. https://doi.org/10.1007/978-94-011-3648-8_40
Download citation
DOI: https://doi.org/10.1007/978-94-011-3648-8_40
Publisher Name: Springer, Dordrecht
Print ISBN: 978-1-85166-678-2
Online ISBN: 978-94-011-3648-8
eBook Packages: Springer Book Archive