Applied Evolutionary Algorithms in Java

Portada
Springer Science & Business Media, 2003 M04 30 - 219 páginas
Genetic algorithms provide a powerful range of methods for solving complex engineering search and optimization algorithms. Their power can also lead to difficulty for new researchers and students who wish to apply such evolution-based methods. "Applied Evolutionary Algorithms in Java" offers a practical, hands-on guide to applying such algorithms to engineering and scientific problems. The concepts are illustrated through clear examples, ranging from simple to more complex problems domains; all based on real-world industrial problems. Examples are taken from image processing, fuzzy-logic control systems, mobile robots, and telecommunication network optimization problems. The Java-based toolkit provides an easy-to-use and essential visual interface, with integrated graphing and analysis tools. Topics and features: *inclusion of a complete Java toolkit for exploring evolutionary algorithms *strong use of visualization techniques, to increase understanding *coverage of all major evolutionary algorithms in common usage *broad range of industrially based example applications *includes examples and an appendix based on fuzzy logic This book is intended for students, researchers, and professionals interested in using evolutionary algorithms in their work. No mathematics beyond basic algebra and Cartesian graphs methods are required, as the aim is to encourage applying the Java toolkit to develop the power of these techniques.
 

Contenido

Introduction to Evolutionary Computing
3
12 History of Evolutionary Computing
4
13 Obstacles to Evolutionary Computation
5
15 Problem Domains
6
152 Optimisation versus Robustness
8
154 Fuzzy Logic
9
155 Bayesian networks
11
156 Artificial Neural Networks
12
Future Directions in Evolutionary Computing
103
63 Speciation and Distributed EA Methods
105
632 Parallel Genetic Programming with Mobile Agents
106
635 EA Visualisation Methods
109
64 Advanced EA Techniques
110
641 Multiobjective Optimisation
111
644 Parameter Control
112
65 Artificial Life and Coevolutionary Algorithms
113

157 Feedforward Networks
13
16 Applications
14
161 Problems
15
17 EvolutionBased Search
16
173 Pascal and Fortran
17
176 ObjectOriented Design
18
18 Summary
19
Further Reading
20
Principles of Natural Evolution
21
212 Biological Genes
22
221 Transcription from DNA to RNA
23
224 No Lamarckianism
24
228 Evolvability
25
2210 Sexual Recombination
26
2213 Dynamics and Morphogenesis
27
Further Reading
28
Genetic Algorithms
29
33 GA Theory
32
331 Deception
33
34 GA Operators
34
342 Crossover
35
343 Multipoint Crossover
36
345 FitnessProportionate Selection
38
346 Disadvantages of FitnessProportionate Selection
39
348 Tournament Selection
40
349 Scaling Methods
41
36 Selecting GA Methods
42
362 Operator Choice
44
38 Summary
47
Further Reading
48
Genetic Programming
49
421 VariableLength and TreeBased Representations
51
424 Function Closure
52
427 Graph Structure Encoding
53
43 GP Operators
54
434 Controlling Genome Growth
55
44 Genetic Programming Implementation
56
45 Summary
57
Further Reading
58
Engineering Examples Using Genetic Algorithms
59
53 Basics of Image Processing
60
532 Lookup Tables
61
54 Java and Image Processing
62
541 Example Application VEGA
63
55 Spectrographic Chromosome representation
69
56 Results
70
57 Summary Evolved Image Processing
72
58 Mobile Robot Control
73
581 Artificial Intelligence and Mobile Robots
74
583 Static Worlds
75
584 Reactive and Bottomup Control
76
585 Advantages of Reactive Control
78
59 Behaviour Management
79
591 Behaviour Synthesis Architecture
80
592 Standard Control Methods PID
83
510 Evolutionary Methods
84
Natural Agents
85
511 Fuzzy Logic Control
86
5111 Fuzzy Control of Subsumption Architectures
89
513 Robot Simulator
91
5131 The Robot Control Architecture
92
5132 Related Work
95
5133 Results
96
514 Analysis
100
515 Summary Evolving Hybrid Systems
101
66 Summary
115
Further Reading
116
The Future of Evolutionary Computing
117
73 Future Directions in Evolutionary Computing
118
732 Biological Inspiration
119
74 Conclusion
121
Bibliography
123
Appendix A
135
A2 CC++ based EA Software
136
A4 Java Reference Guides
138
Appendix B
139
B21 Vectors and Arraylists
149
B3 Application Design
150
An Evolutionary and Ecosystem Research Platform
151
B41 Introduction
152
B43 Key Classes
154
B44 Configuration
155
B45 Illustrative Example Systems
156
B47 An Ecosystem Simulation Based on Echo
157
B48 Coevolutionary Function Optimisation
158
B49 Telecommunications Research Using Eos
159
A Telecommunications Application
160
B5 Traveling Salesman Problem
161
B6 Genetic Programming
165
B61 Observations from Running GPsys on the Lawnmower Problem
166
Eos References
169
Fuzzy Logic Systems
171
C2 Fuzzy Set Theory
172
C21 Fuzzy Operators
173
C23 Fuzzy IF
174
C24 Fuzzy Associative Memories
175
C25 Fuzzy Control systems
176
C26 Defuzzification
177
C27 Fuzzy Applications
180
C31 Advantages of Fuzzy Systems
181
C4 Summary
182
Appendix D
183
Programming Language and RunTime Environment
184
Units of Measure
185
NetworkLocal Connections Versus Dynamically Loaded Clients
186
Why a ClientServer Architecture?
187
Network and Local Connection Issues
188
Communication via Events and Requests
189
Keeping the RP1 Protocol LanguageIndependent
190
The port and hostName Properties
191
Loading RsProperties Files as a Resource
192
The Server
193
The Scheduler
194
The FloorPlan
196
Building a Virtual Robot
198
Life Cycle of the Demonstration Clients
199
How ClnMain Extends RsClient and Implements RsRunnable
201
The RsRunnable Interface
202
Building RsRunnable and RsClient into ClnMain
203
DemoMain Implements RsRunnable But Does Not Extend RsClient
204
Uploading the Body Plan
206
Running the Event Loop
207
Physical Layout of ClientZero
209
RsBodyShape
211
RsWheelSystem
212
The Sensor Classes
214
RsBodyTargetSensor
215
Events and Requests
216
Index
218
Derechos de autor

Otras ediciones - Ver todas

Términos y frases comunes

Información bibliográfica