Intelligent Systems: Fusion, Tracking and Control (CSI, Control and Signal Image Processing Series)

Team DDU......Page 1
Editorial Foreword......Page 6
Preface......Page 8
Acknowledgements......Page 10
Contents......Page 12
List of Figures......Page 20
List of Tables......Page 24
I......Page 28
1.1.1 Will an intelligent system match the human intelligence?......Page 30
1.2.1 Examples of existing intelligent systems......Page 31
1.2.2 Learning from nature......Page 32
1.2.3 How do we qualify a system as intelligent?......Page 33
1.3 What Makes an Intelligent System......Page 34
1.3.1 How an intelligent system sees the world?......Page 35
1.4 Challenges of Building Intelligent Systems......Page 36
2.1 Introduction......Page 38
2.2.2 Relationships......Page 39
2.2.2.2 Aggregation......Page 40
2.2.2.4 Generalization......Page 41
2.3 Data Structure and Storage......Page 42
2.3.1 Data structure......Page 43
2.4 What is a Software Agent-based System?......Page 44
2.4.1 What is an agent?......Page 45
2.4.2.2 Multi-agent negotiation......Page 46
2.5 An Agent-Oriented Software Development Methodology......Page 47
2.5.1 Analysis......Page 49
2.6 Object- and Agent-Oriented Software Engineering Issues......Page 50
3.1 Introduction......Page 54
3.2.1 Biological inspiration......Page 55
3.2.3.1 Introduction to GAs......Page 56
3.2.3.2 Operator......Page 58
3.2.3.3 Parameters of GAs......Page 59
3.2.3.5 Further information on crossover and mutation operators......Page 60
3.3.1 Biological inspiration......Page 61
3.3.2 Historical perspective......Page 62
3.3.3.1 Basic units......Page 64
3.3.3.2 Network topology......Page 65
3.3.4.1 Supervized learning......Page 68
3.3.4.2 First order gradient methods......Page 70
3.3.4.3 Reinforcement learning......Page 75
3.3.4.4 Unsupervized learning......Page 77
3.3.5 Training of neural networks......Page 79
3.4.3 Basic elements of fuzzy systems......Page 80
3.4.3.1 Fuzzification......Page 81
3.4.3.2 Fuzzy lingusitic inference......Page 82
3.4.3.3 Defuzzification......Page 83
3.5 Expert Systems......Page 84
3.6 Fusion of Biologically Inspired Algorithms......Page 86
3.7 Conclusions......Page 87
4.1 Introduction......Page 88
4.2 Classifying Types of Sensor Systems......Page 89
4.3 Sensing Technology in Perspective......Page 90
4.3.1.1 Imagery radar......Page 93
4.3.1.5 Laser radar......Page 94
4.3.3 Acoustic and sonar sensor systems......Page 95
4.3.4.2 Indirect measurement......Page 96
4.3.5.1 Gas sensors......Page 97
4.4 Sensor System Design Issues......Page 98
4.4.2 Sensing platforms......Page 99
II......Page 102
5.1 Data and Information Fusion System - What and Why......Page 104
5.2 What are the Different Levels of Fusion Process?......Page 107
5.3 Issues on Fusing Multiple Sources......Page 111
5.4.1 How to make sense of information?......Page 115
5.4.2 Memory structure for cognitive intelligence processing......Page 117
5.5.1 Centralized fusion architecture......Page 118
5.5.4 Hierarchical fusion and hybrid fusion architecture......Page 119
5.5.5 Is there a difference between decentralized and distributed fusion?......Page 120
5.5.6 Is there a single best architecture?......Page 121
5.6.1.1 Data association and correlation......Page 122
5.6.1.2 State estimation......Page 124
5.6.2 Level 2 and 3 - Situation and impact assessment......Page 125
5.6.2.1 Clustering and aggregation......Page 126
5.6.2.2 Learning, reasoning and knowledge embedding and retrieval strategies......Page 127
5.7 Complementary and Competitive Data......Page 129
5.8 Summary......Page 130
6.1 Why Data Association......Page 134
6.2 Methods of Data Association......Page 135
6.2.1 Two-stage data association process......Page 136
6.3 Nearest Neighbour......Page 137
6.4.1 Multiple hypothesis tracking......Page 138
6.4.1.1 Measurement-oriented approach......Page 139
6.5 Probabilistic Data Association......Page 140
6.5.1 Probabilistic data association filter (PDAF)......Page 141
6.5.2 Joint probabilistic data association filter (JPDAF)......Page 144
6.6 Graph-Based Data Association......Page 146
6.6.1.1 Association strength......Page 147
6.6.2.1 The reduction algorithms......Page 148
6.7.1 Neural data association methods......Page 150
6.7.2 Fuzzy logic and knowledge-based data association methods......Page 152
7.1 Introduction......Page 154
7.2 Basics of a Tracking Process......Page 156
7.3 Problem Formulation......Page 157
7.4 Fixed-gain Filter......Page 158
7.4.1 What are the criteria for selecting the fixed gain?......Page 160
7.5 Kalman Filter......Page 161
7.6 Multiple Model Approach......Page 163
7.6.1.1 Input mixer (Interaction)......Page 164
7.6.1.2 Filtering updates......Page 165
7.6.1.4 Output mixer......Page 166
7.6.2.2 Filters......Page 170
7.6.2.3 Results and analysis......Page 171
7.6.2.4 Summary......Page 172
7.6.3 Comparison between IMM and the generic particle filter......Page 173
7.7.1 Coordinates......Page 174
7.7.2 Gating issues......Page 175
7.8 Conclusions......Page 176
8.1 What is Cooperative Intelligence......Page 178
8.1.1 Characteristics of cooperative intelligence......Page 179
8.2.1 Network of nodes......Page 180
8.2.2 Decentralized tracking......Page 181
8.3 Measurement and Information Fusion......Page 182
8.3.1 Decentralized Kalman filter......Page 183
8.3.1.1 Formulation of a global state estimate......Page 184
8.3.2 Information filter......Page 186
8.3.2.1 Implementation of the decentralized information filter equation......Page 188
8.3.3 Measurement Fusion......Page 189
8.4 Track or State Vector Fusion......Page 190
8.4.1 Convex combination......Page 191
8.4.2 Covariance intersection......Page 192
8.4.4 Bar-Shalom and Campo combination......Page 193
8.4.5 Information de-correlation......Page 194
8.5.1 Target and sensor modelling......Page 195
8.5.3 Experiment results......Page 196
8.6 Summary......Page 200
9.1 Introduction......Page 202
9.2 Multi-level Sensor Management Based on Functionality......Page 204
9.3 Sensor Management Techniques......Page 206
9.3.1 Scheduling techniques......Page 207
9.3.2.2 Descriptive technique......Page 209
9.3.2.4 Other decision-making tools......Page 210
9.4 Sensor Management - Controller and Cueing......Page 211
9.5.1 Fuzzy controller......Page 212
9.5.3 Example 2......Page 214
9.7 Conclusions......Page 215
A.1.2 Constant velocity model......Page 222
A.1.3 Acceleration models......Page 223
A.1.4 Exponentially correlated velocity (ECV) model......Page 224
A.1.5 Exponentially correlated acceleration (ECA) model......Page 225
A.2 Generic Particle Filter......Page 226
A.2.1 Generic particle filter algorithm......Page 227
A.3 Bayesian Classifier......Page 230
A.4 Dempster-Shafer Classifier......Page 232
A.5.1 B-cells or B- lymphocytes......Page 235
A.5.2 T-cell or T-lymphocytes......Page 236
A.5.3 What can we learn from the immune system?......Page 237
A.6.2 Components of ontologies......Page 238
A.6.4 Principles of ontology design......Page 240
A.6.5 Conclusion......Page 241
Bibliography......Page 242
Index......Page 260