Mechatronics is the fusion of mechanics and electronics in the design of intelligent machines. This textbook is concerned with the concepts and techniques of artificial intelligence needed for the design of machines with advanced intelligent behaviour. It explores the topics of pattern recognition, neural networks, scheduling, reasoning, fuzzy logic, rule-based systems, machine learning, control and computer vision.This student guide shows how fifty years of research into artificial intelligence (AI) have borne fruit in the design of better and more intelligent machines. The twin objectives of the text are: to explain the theory of the mainstream ideas of AI and to show how these ideas can be applied in practical engineering situations.
Author(s): Jeffrey Johnson, Philip Picton
Series: Mechatronics, Designing Intelligent Machines, Vol 2 v. 2
Publisher: Newnes
Year: 1995
Language: English
Pages: 376
Front Cover......Page 1
Concepts in Artificial Intelligence......Page 4
Copyright Page......Page 5
Contents......Page 8
Preface......Page 6
Overview of Volume 2......Page 15
1.1 Artificial intelligence in engineering......Page 18
7.3 Why build intelligence into machines?......Page 19
1.4 How much intelligence can be built into machines?......Page 20
1.5 What is artificial intelligence?......Page 21
1.6 How is AI applied to engineering in practice?......Page 23
1.7 The principles behind the applications......Page 25
2.1 Introduction......Page 26
2.2 Theoretical foun dations......Page 32
2.3 Relational patterns and graph matching......Page 35
2.4 Hierarchical structure in pattern recognition......Page 40
2.5 Data transformation in pattern recognition......Page 41
2.6 Pattern recognition using multidimensional data......Page 45
2.7 Multiple classifications and fuzzy sets......Page 61
2.9 Rigorous procedures for training pattern recognizers......Page 63
2.10 Conclusion......Page 67
3.1 Introduction......Page 70
3.2 Tree search......Page 75
3.3 Calculus-based search......Page 85
3.4 Probabilistic search......Page 95
3.5 Conclusion......Page 109
4.1 Introduction......Page 112
4.2 The artificial neural unit......Page 117
4.3 Pattern classification......Page 122
4.4 Feedforward networks......Page 127
4.5 Learning in neural networks......Page 130
4.6 Feedback networks......Page 140
4.7 Uses of the multilayer perceptron......Page 141
4.8 Conclusion......Page 152
5.1 Introduction......Page 154
5.2 Representation in scheduling......Page 155
5.3 Graphs and networks for representing scheduling problems......Page 156
5.4 Shortest paths......Page 159
5.5 Critical path analysis......Page 161
5.6 Critical path activity scheduling......Page 168
5.7 The 'travelling salesman problem'......Page 172
5.8 Intelligent scheduling......Page 187
5.9 Conclusion......Page 189
6.1 lntroduction......Page 192
6.2 Reasoning with certainty......Page 197
6.3 Reasoning with uncertainty......Page 208
6.4 Conclusion......Page 233
7.1 Knowledge-based, rule-based and expert systems......Page 234
7.2 Implementation......Page 256
7.3 Confidence levels and fuzzy rules......Page 263
7.4 Programming language and rule-based system shells......Page 264
7.5 Conclusion......Page 265
8.1 Introduction......Page 266
8.2 Learning by memory......Page 267
8.3 Learning by updating parameters......Page 268
8.4 learning during execution using Bayesian updating......Page 269
8.5 learning from examples......Page 275
8.6 learning by analogy......Page 284
8.7 Learning by discovery......Page 286
8.8 Conclusion......Page 289
9.1 Introduction......Page 290
9.2 The broom-balancer......Page 291
9.3 Classical solution......Page 294
9.4 Neural network solution......Page 302
9.5 Genetic algorithms......Page 308
9.6 Fuzzy rules......Page 310
9.7 Hierarchical control of complex systems......Page 320
9.8 Conclusion: principles for intelligent control design......Page 328
10.1 Introduction......Page 332
10.2 Abstracting information from digital images......Page 334
10.3 The nature of digital images......Page 339
10.4 Computer vision versus computer graphics......Page 349
10.5 Object recognition and measurement......Page 350
10.6 A summary of the basic techniques in computer vision......Page 362
10.7 A hierarchical architecture for computer vision......Page 368
10.8 Conclusion: computer vision in intelligent machines......Page 370
11.1 An introduction to blackboard systems......Page 372
11.2 The blackboard system as a development environment......Page 373
11.3 Running many rule-based systems in parallel......Page 374
11.5 Implementing a perception subsystem......Page 376
11.6 Implementing a cognition subsystem......Page 377
11.8 Integration: emergent behaviour and control......Page 379
11.9 Blackboard systems and the concepts and techniques of AI......Page 381
11.10 Conclusion......Page 384
Acknowledgements......Page 385
Index......Page 386
