This book is dedicated for engineers and researchers who would like to increase the knowledge in area of mobile mapping systems. Therefore, the flow of the derived information is divided into subproblems corresponding to certain mobile mapping data and related observations’ equations. The proposed methodology is not fulfilling all SLAM aspects evident in the literature, but it is based on the experience within the context of the pragmatic and realistic applications. Thus, it can be supportive information for those who are familiar with SLAM and would like to have broader overview in the subject.
The novelty is a complete and interdisciplinary methodology for large-scale mobile mapping applications. The contribution is a set of programming examples available as supportive complementary material for this book. All observation equations are implemented, and for each, the programming example is provided. The programming examples are simple C++ implementations that can be elaborated by students or engineers; therefore, the experience in coding is not mandatory.
Moreover, since the implementation does not require many additional external programming libraries, it can be easily integrated with any mobile mapping framework. Finally, the purpose of this book is to collect all necessary observation equations and solvers to build computational system capable providing large-scale maps.
Author(s): Janusz Będkowski
Series: Cognitive Intelligence and Robotics
Publisher: Springer
Year: 2022
Language: English
Pages: 314
City: Singapore
Preface
Acknowledgements
Contents
Acronyms
Symbols
Part I Introduction
1 Introduction
1.1 Novelty and Contribution
1.2 Terminology
1.2.1 Basic Terms
References
2 Mobile Mapping Systems
2.1 Commercial Measurement Instruments
2.2 Mobile Mapping Data
2.2.1 IMU
2.2.2 Odometry
2.2.3 Light Detection and Ranging
2.2.4 Mono-, Stereo-, Spherical and Depth Cameras
2.3 Ground Truth Data Sources
2.3.1 Large-scale Data Sets
2.3.2 Long-Term Data Sets
2.3.3 Terrestrial Laser Scanning Data
References
Part II Methodology
3 Coordinate Systems
3.1 Introduction
3.2 Rotations
3.2.1 Tait–Bryan and Euler Angles
3.2.2 Euler–Rodrigues Formula
3.2.3 Quaternions
3.3 Cartesian Coordinate System
3.4 Spherical Coordinate System
3.5 Geographic Coordinate System
3.5.1 Gauss–Krüger Geographic Coordinate System and UTM
References
4 Simultaneous Localization and Mapping
4.1 Introduction
4.2 Weighted Non-linear Least Squares Optimization
4.2.1 Introduction
4.2.2 Solvers
4.2.3 Robust Least Squares
4.2.4 Manifolds
4.3 Probabilistic Formulation
4.4 Graphical Representation with Factor Graphs
4.5 Graph SLAM
4.5.1 Pose Graph SLAM
4.6 Approach
References
5 Trajectory Estimation
5.1 Motion Model
5.2 Kalman Fiter
5.2.1 Extended Kalman Filter
5.3 Particle Filter
5.3.1 Variable of Interest
5.3.2 Prediction Stage (Motion Model)
5.3.3 Use of Semantic Data in the Particle Filter
5.3.4 Update Stage and Resampling
5.4 Structure from Motion
5.4.1 Pinhole Camera Model
5.4.2 Essential and Fundamental Matrices
5.4.3 Homography Matrix
5.4.4 Bundle of Rays Intersection (Triangulation)
5.4.5 Bundle Adjustment
5.5 Lidar Odometry
References
6 Errors
6.1 Uncertainty, Accuracy and Precision
6.2 Evaluation Metrics
6.3 Mean and Covariance of the Point Cloud
6.4 Black-Box Method
6.5 Error Propagation Law
6.6 Closed-Form Estimate of Covariance
6.7 Global Positioning System Accuracy Assessment
References
Part III Observation Equations
7 Camera Metrics
7.1 Pinhole Camera Model
7.1.1 Pinhole Camera Reprojection Error
7.1.2 Intrinsic Calibration of Perspective Camera
7.1.3 Calculating Pinhole Camera External Orientation Using Plücker Lines
7.2 Metric Camera Model
7.2.1 Colinearity Observation Equation
7.2.2 Coplanarity Observation Equation
7.3 Equirectangular Camera Model
7.3.1 Colinearity Observation Equation
7.3.2 Coplanarity Observation Equation
References
8 LiDAR Metrics
8.1 Point to Point Metrics
8.1.1 Point to Point
8.1.2 Point to Point—Source to Target
8.1.3 Point to Point—Source to Landmark
8.1.4 Point to Point with Scale
8.1.5 Semantic Point to Point
8.1.6 Point to Projection
8.2 Point to Feature Metrics
8.2.1 Point to Line
8.2.2 Point to Plane
8.2.3 Distance Point to Plane
8.2.4 Point to Surface
8.3 Feature to Feature Metrics
8.3.1 Line to Line
8.3.2 Plane to Plane
8.4 Reflection Observation Equation
8.5 Normal Distributions Transform
8.6 Local Geometric Features—Surfels
8.7 Nearest Observations Search
8.7.1 Anisotropic Nearest Neighborhood Search
References
9 Constraints
9.1 Norm of the Quaternion
9.2 Fixed Optimized Parameter
9.3 Anisotropic Motion
9.4 Linear and Square Functions
9.5 Relative Pose
9.5.1 Variant (1)
9.5.2 Variant (2)
9.6 Smoothness
9.7 Georeference
9.7.1 Case 1: Georeferenced Position (tx,ty,tz)
9.7.2 Case 2: Georeferenced Pose [R,t]
9.7.3 Case 3: Georeferenced Primitive (LiDAR Metric)
9.7.4 Case 4: Georeferenced Primitive (Metric Camera)
9.8 Distance to Circle
10 Metrics' Fusion
10.1 Introduction
10.2 Rectangular Object with Unknown Width/Height
10.3 Sheaf of Planes Intersection
10.4 Surface Reconstruction from LiDAR Point Cloud
10.5 Multi-measurement Instrument System
10.5.1 Multi-sensor Extrinsic Calibration
References
Part IV Applications
11 Building Large-Scale SLAM
11.1 Normal Equation for BA
11.2 Trajectory Centric Versus Map-Centric Approach
11.3 Continuous Trajectory SLAM
11.4 Design Matrix—Programming Example
11.5 Loop Closing and Change Detection
11.5.1 Processing Video Information
11.5.2 Processing LiDAR Information
References
12 Final Map Qualitative and Quantitative Evaluation
12.1 Measurement Instrument
12.2 TLS Matching
12.3 Mobile Backpack Mapping System Qualitative Evaluation
12.4 Improving State-of-the-Art Visual SLAM
Reference
Appendix Appendix
Reference
