Developments in Soft Computing

235 Pages·2001·11.088 MB·English

by  Kevin Warwick (auth.), Robert John, Ralph Birkenhead (eds.)

Developments in Soft Computing
Advances in Soft Computing
Editor-in-chief
Prof. Janusz Kacprzyk
Systems Research Institute
Polish Academy of Sciences
ul. Newelska 6
01-447 Warsaw, Poland
E-mail: [email protected]
http://www.springer.de/cgi-binlsearch-bock.pl?series=4240
Esko Turunen
Mathematics Behind Fuzzy Logic
1999. ISBN 3-7908-1221-8
Robert Fuller
Introduction 10 Neuro-Fuzzy Systems
2000. ISBN 3-7908-1256-0
Robert John and Ralph Birkenhead (Eds.)
Soft Computing Techniques
and Applications
2000. ISBN 3-7908-1257-9
Mieczyslaw Klopotek, Maciej Michalewicz
and Slawomir T. WierzchOll (Eds.)
Intelligent Information Systems
2000. ISBN 3-7908-1309-5
Peter Sincak, Jan Vascak, Vladimir Kvasnicka
and Radko Mesiar (Eds.)
The State of the Art in Computational Intelligence
2000. ISBN 3-7908-1322-2
Bernd Reusch, Karl-Heinz Temme (Eds.)
Computational Intelligence in Theory and Practice
2001. ISBN 3-7908-1357-5
Robert John· Ralph Birkenhead
Editors
Developments
in Soft Computing
With 75 Figures
and 28 Tables
Springer-Verlag Berlin Heidelberg GmbH
Robert John
Ralph Birkenhead
De Montfort University
Department of Computer Science
The Gateway
Leicester LEI 9BH
United Kingdom
[email protected]
[email protected]
ISSN 1615-3871
ISBN 978-3-7908-1361-6
Cataloging-in-Publication Data applied for
Die Deutsche Bibliothek - CIP-Einheitsaufnahme
Developments in soft computing: with 28 tables / Robert John; Ralph Birkenhead (ed.). - Heidelberg; New
York: Physica-Verl., 2001
(Advances in soft computing)
ISBN 978-3-7908-1361-6 ISBN 978-3-7908-1829-1 (eBook)
DOI 10.1007/978-3-7908-1829-1
This work is subject to copyright. Ali rights are reserved, whether the whole or part of the material is con
cemed, specifically the rights of translation, reprinting, reu se of illustrations, recitation, broadcasting, repro
duction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts
thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its
current version, Violations are liable for prosecution under the German Copyright Law.
© Springer-Verlag Berlin Heidelberg 2001
Originally published by Physica-Verlag Heidelberg New York in 2001
The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply,
even in the absence of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore frec for general use.
Hardcover Design: Erich Kirchner, Heidelberg
SPIN 10789622 88/2202-5 4 3 2 l O -Printed on acid-free paper
Preface
Soft Computing has come of age. In particular, Artificial Neural Networks, Fuzzy
Logic and Evolutionary Computing now play an important role in many domains
where traditional techniques have been found wanting. As this volume confirms,
hybrid solutions that combine more than one of the Soft Computing approaches
are particularly successful in many problem areas. This volume contains twenty
seven papers that represent the activities of a number of leading researchers in
Soft Computing who presented their work at the International Conference on
Recent Advances in Soft Computing 2000 at De Montfort University in Leicester.
We were fortunate in that the conference coincided with a workshop on mobile
robots and we were able to have a joint session on areas of mutual interest to the
n'lo groups of participants. We are particularly pleased that the volume includes a
paper presented to that session by Professor Kevin Warwick, in which he gives
some interesting views on the future of intelligent mobile robots and their
relationship with the human race.
We would also like to thank the other writers and plenary speakers for their
contributions which provide an insight into current research activities from both a
theoretical and application view. We hope you find the volume interesting and, if
you are a researcher, that the work here will inform your own research.
Robert John and Ralph Birkenhead
Centre for Computational Intelligence
De Montfort University
Leicester
United Kingdom
Contents
Preface
Plenary Address
How Intelligent Can a Mobile Robot Be? 3
Kevin Warwick
Part 1 Theoretical Development
Hybrid Heuristics for Optimal Design of Artificial Neural Networks 15
Ajith Abraham and Baikunth Nath
Efficient Image Sequence Analysis Using Fuzzy Techniques 23
M.J.Allen, Q.H.Mehdi, N.E.Gough and I.M.Coulson
Automatic Generation of Fuzzy Rule-based Models from Data by Genetic Algorithms 31
P.P.Angelov, R.A.Buswell, V.I. Hanby and JA. Wright
Resemblance Relations and Fuzzy Weak Orderings 41
D.Boixader and JRecasens
Fuzzy Control System with B-operations 48
Bohdan S.Butkiewicz
Roughness Indicator Fuzzy Set 56
Kankana Chakrabarty
VIII
Dynamic Recognition States for Chaotic Neural Networks 62
Nigel Crook and Tjeerd Olde Scheper
A Neural Network Phoneme Classification Based on Wavelet Features 73
o.Farooq and S.Datta
Fuzzy Symptoms and a Decision Support Index for the Early Diagnosis of 79
Confusable Diseases
P. R.In nocent
Finding Diverse Examples Using Genetic Algorithms 92
Colin GJohnson
Evolving User Profiles to Reduce Internet Information Overload 100
John Pagonis and Mark C.Sinclair
Learning Fuzzy Classification Rules from Data 108
Hans Roubos. Magne Setnes and Janos Abonyi
Increasing Diversity in Genetic Algorithms 116
Tim Watson and Peter Messer
Modelling Non-Numeric Linguistic Variables 124
Jon Williams. Nigel Steele and Helen Robinson
Part 2 Application Exemplars
A Preliminary Fuzzy Model to Identify Abnormal Cervical Smears Using Fourier 135
Transform Infrared Spectroscopy
Robert Adama-Acquah. Jonathan Garibaldi and Ian Symonds
Application of Soft-Computing Techniques in Modelling of Buildings 143
D.Azzi. A.E.Gegov. G.s. Virk .B.P.Haynes and K.Alkadhimi
Neuro-Fuzzy Adaptive Strategies Applied to Power System Stabilization 151
Walter Barra Junior and Jose Augusto Lima Barreiros
IX
Using Genetic Algorithms to Incorporate Taste into Diets 159
S.BerIY and V.Lowndes
The Application of Fuzzy Logic and Genetic Algorithms to Oil Exploration 167
S.J. Cuddy and P. w.J. Glover
Symbolic and Numerical Regression: Experiments and Applications 175
J. W.Davidson, D.A.Savic and G.A. Walters
G.A. Optimisation of PID Controllers - Optimal Fitness Functions 183
D.Drabble, P. v.s.Ponnapalli and M. Thomson
Fuzzy Logic for Behaviour Co-ordination and Multi-Agent Formation in RoboCup 191
Hakan Duman and Hosheng Hu
A Hierarchical Fuzzy Genetic Multi-Agent Architecture for Intelligent Buildings 199
Sensing and Control
Hani Hagras, Victor Callaghan, Martin Colley and Graham Clarke
The Use of Evolutionary and Fuzzy Models for Oncological Prognosis 207
A.A.Odusanya, M.o.Odetayo, D.Petrovic and R.N.G.Naguib
Evolving Protein Similarity Scoring Matrices Using Differential Evolution 216
Ketan Patel, Andrew Tuson, Andrew Coulson, Shane Sturrock and Robert Fisher
A Machine Process Cost Estimate Utilising Genetic Algorithms 224
J. G. Qi, D.J.Stockton, D.K.Harrison and J.Ardon-Finch
Plenary Address
How Intelligent Can a Mobile Robot Be?
Kevin Warwick
Dept. of Cybernetics, University of Reading, Whiteknights,
Reading, RG6 6AY
Summary:
A look is taken at the world of mobile robots. An argument is given for robots
being not only intelligent but also conscious. Cyborg type robots are considered
and the robots presently operating in the Science Museum, London are
introduced
1 Introduction
Firstly let us establish some ground rules. A robot can be alive. A robot can be
conscious. A robot can be intelligent. But how intelligent can a robot be?
The whole concept of intelligence has been a major philosophical discussion piece
for thousands of years. Even folk such as Plato and Cicero voiced their opinion on
the topic. In the last century the debate continued and over the 40 or so years
from 1960 it received a concept injection from the new field of Artificial
Intelligence. When this area itself bursts into life, immediately the question was
raised, what exactly is artificial intelligence? Perhaps the most famous and oft
quoted answer was that Minsky (1968), 'Artificial Intelligence is the science of
making machines do things that would require intelligence if done by men'. Over
the years this definition has been fleshed out, e.g. by Boden (1987), but basically
it has remained the same.
The Minsky definition has two major problems associated with it. Firstly it does
not actually tackle the question of What is Intelligence? Secondly it links
intelligence solely as a human property with Artificial Intelligence (AI) as a sort
of mimicking or copying exercise. I personally do not wish to dispute the fact that
humans are, by and large, intelligent in their way. However I feel that other
creatures on earth are also intelligent, each in their way. Despite this the Minsky
definition still seems to have entrenched itself as a basic way of life. For example,
R. John et al. (eds.), Developments in Soft Computing
© Springer-Verlag Berlin Heidelberg 2001