Parallel and Distributed Map Merging and Localization (eBook)

Preface 6
Contents 8
1 Introduction 10
1.1 Motivation 10
1.2 Classical Approaches 14
1.3 Document Organization 16
References 17
2 Distributed Data Association 20
2.1 Introduction 20
2.2 Problem Description 22
2.2.1 Matching Between Two Cameras 23
2.2.2 Centralized Matching Between n Cameras 23
2.2.3 Distributed Matching Between n Cameras 24
2.3 Propagation of Local Associations 25
2.4 Algorithm Based on Trees 28
2.5 Feature Labeling 32
2.6 Algorithm Based on the Maximum Error Cut 35
2.7 Simulations 41
2.8 Closure 44
References 44
3 Distributed Localization 46
3.1 Introduction 46
3.2 Problem Description 48
3.3 Planar Localization from Noisy Measurements 50
3.3.1 Centralized Algorithm 50
3.3.2 Distributed Algorithm 52
3.4 Centroid-Based Position Estimation from Noisy Measurements 58
3.4.1 Position Estimation Relative to an Anchor 59
3.4.2 Centralized Centroid-Based Position Estimation 62
3.4.3 Distributed Centroid-Based Position Estimation 64
3.5 Simulations 67
3.6 Closure 71
References 71
4 Map Merging 74
4.1 Introduction 74
4.2 Problem Description 78
4.3 Dynamic Map Merging Algorithm 80
4.4 Properties of the Dynamic Map Merging Algorithm 84
4.5 Simulations 92
4.6 Closure 94
References 94
5 Real Experiments 97
5.1 Data Association with Visual Data 97
5.2 Map Merging with Visual Data 100
5.3 Data Association, Localization, and Map Merging with RGB-D 105
5.4 Closure 113
References 113
6 Conclusions 114
Appendix AAveraging Algorithms and MetropolisWeights 116
Appendix BAuxiliary Results for Distributed Localization 118
Index 121