We recently reported on the application of a machine-learning (ML) technique to automated flight control using a simulated F-16 combat plane (Michie and Camacho 1994). Subsequent tests of our data-induced flying model have broadly confirmed the reported results but have also identified a lack of robustness. We had underestimated the latter and now regard our report (Michie and Camacho 1994) as being, by omission, potentially misleading.

The goal of this research is to learn to control the attitude of an orbiting satellite. To this end, we are investigating the possibility of using adaptive controllers for such tasks. Laboratory tests have suggested that rule-based methods can be more robust than systems developed using traditional control theory. The BOXES learning system, which has already met with success in simulated laboratory tasks, is an effective design framework for this new exercise.

This article reports on experiments performed using a black-box simulation of a spacecraft. The goal of this research is to learn to control the attitude of an orbiting satellite. The space-craft must be able to operate with minimal human supervision. To this end, we are investigating the possibility of using adaptive controllers for such tasks. Laboratory tests have suggested that rule-based methods can be more robust than systems developed using traditional control theory. The BOXES learning system, which has already met with success in simulated laboratory tasks, is an effective design framework for this new exercise.

Consider a class of computing problem for which all bananas is left as an exercise for the reader, or the sufficiently short programs are too slow and all sufficiently monkey. When it has been possible to couple causal models problems of this kind were left strictly alone for the first with various kinds and combinations of search, twenty-years or so of the computing era. There were two mathematical programming and analytic methods, then good reasons. First, the above definition rules out both evaluation of t has been taken as the basis for "high road" the algorithmic and the database type of solution. In "low road" representations Second, in a pinch, a human expert could usually be s may be represented directly in machine memory as a set found who was able at least to compute acceptable A recent pattern-directed allocation, inventory optimisation, or whatever large heuristic model used for industrial monitoring and control combinatorial domain might happen to be involved.

Edinburgh: Edinburgh University Press

Note: PDF of full volume downloadable by clicking on title above (32.8 MB). Selected individual chapters available from the links below.CONTENTSINTRODUCTORY MATERIALMATHEMATICAL FOUNDATIONS1 Program scheme equivalences and second-order logic. D. C. COOPER 32 Programs and their proofs: an algebraic approach.R. M. BURSTALL and P. J. LANDIN 173 Towards the unique decomposition of graphs. C. R. SNOW andH. I. SCOINS 45THEOREM PROVING4 Advances and problems in mechanical proof procedures. D. PRAWITZ 595 Theorem-provers combining model elimination and Tesolution.D. W. LOVELAND 736 Semantic trees in automatic theorem-proving. R. KOWALSKI andP. J. HAYES 877 A machine-oriented logic incorporating the equality relation.E. E. SIBERT 1038 Paramodulation and theorem-proving in first-order theories withequality. G. ROBINSON and L. Wos 1359 Mechanizing higher-order logic. J. A. ROBINSON 151DEDUCTIVE INFORMATION RETRIEVAL10 Theorem proving and information retrieval. J. L. DARLINGTON 17311 Theorem-proving by resolution as a basis for question-answeringsystems. C. CORDELL GREEN 183MACHINE LEARNING AND HEURISTIC PROGRAMMING12 Heuristic dendral: a program for generating explanatory hypothesesin organic chemistry. B. BUCHANAN, G. SUTHERLAND andE. A. FEIGENBAUM 20913 A chess-playing program. J. J. SCOTT 25514 Analysis of the machine chess game. I. J. GOOD 26715 PROSE—Parsing Recogniser Outputting Sentences in English.D. B. VIGOR, D. URQUHART and A. WILKINSON 27116 The organization of interaction in collectives of automata. 285V. I. VARSHAVSKY COGNITIVE PROCESSES: METHODS AND MODELS17 Steps towards a model of word selection. G. R. Kiss 31518 The game of hare and hounds and the statistical study of literaryvocabulary. S. H. STOREY and M. A. MAYBREY 33719 The holophone —recent developments. D. J. WILLSHAW andH. C. LONGUET-HIGGINS 349PATTERN RECOGNITION20 Pictorial relationships — a syntactic approach. M. B. CLOWES 36121 On the construction of an efficient feature space for optical characterrecognition. A. W. M. COOMBS 38522 Linear skeletons from square cupboards. C. J. HILDITCH 403PROBLEM-ORIENTED LANGUAGES23 Absys 1: an incremental compiler for assertions; an introduction.J. M. FOSTER and E. W. ELCOCK 423PRINCIPLES FOR DESIGNING INTELLIGENT ROBOTS24 Planning and generalisation in an automaton/environment system.J. E. DORAN 43325 Freddy in toyland. R. J. POPPLESTONE 45526 Some philosophical problems from the standpoint of artificialintelligence. J. MCCARTHY and P. J. HAYES 463INDEX 505 Machine Intelligence Workshop