Robot Mapping - WS 2015/16 (original) (raw)

Info

Robot Mapping

What is this lecture about?

The problem of learning maps is an important problem in mobile robotics. Models of the environment are needed for a series of applications such as transportation, cleaning, rescue, and various other service robotic tasks. Learning maps requires solutions to two tasks, mapping and localization. Mapping is the problem of integrating the information gathered with the robot's sensors into a given representation. It can intuitively be described by the question ``What does the world look like?'' Central aspects in mapping are the representation of the environment and the interpretation of sensor data. In contrast to this, localization is the problem of estimating the pose of the robot relative to a map. In other words, the robot has to answer the question "Where am I?" These two tasks cannot be solved independently of each other. Solving both problems jointly is often referred to as the simultaneous localization and mapping (SLAM) problem. There are several variants of the SLAM problem including passive and active approaches, topological and metric SLAM, feature-based vs. volumetric approaches, and may others.

The lecture will cover different topics and techniques in the context of environment modeling with mobile robots. We will cover techniques such as SLAM with the family of Kalman filters, information filters, particle filters. We will furthermore investigate graph-based approaches, least-squares error minimization, techniques for place recognition and appearance-based mapping, and data association. The exercises and homework assignments will also cover practical hands-on experience with mapping techniques, as basic implementations will be part of the homework assignments.

Organization

Schedule

Date Topic Slides Recordings Homework Assignment Relevant Literature
19.10. Course Introduction 00.pdf MP4 --- ---
19.10. Introduction to Robot Mapping 01.pdf MP4 --- Springer Handbook on Robotics Chapter 37.1 + 37.2 Probabilistic Robotics Chpater 10.1
22.10. Homogenous Coordinates 02.pdf MP4 Sheet1 Octave-Code Octave Help Wiki page
26.10. Bayes Filter and Related Models (Summary of Introduction to Mobile Robotics) 03.pdf MP4 Sheet2 Introduction to Mobile Robotics lecture Probabilistic Robotics Book, Chapters 2, 5, 6
02.11. Kalman Filter and Extended Kalman Filter (EKF) 04.pdf MP4 (MP4 WS 2014) Sheet3 Probabilistic Robotics Book, Chapter 3.1-3.3 Manipulating the Multivariate Gaussian Density Kalman Filter Tutorial by Welch and Bishop Notes on Univariate Gaussian Distributions and One-Dimensional Kalman Filters
02.11. EKF SLAM 05.pdf MP4 Probabilistic Robotics Book, Chapter 10
09.11. Unscented Kalman Filter (UKF) 06.pdf MP4 Sheet4 Octave-Code Probabilistic Robotics Book, Chapter 3.4
09.11. Extended Information Filter (EIF) 07.pdf Probabilistic Robotics Book, Chapter 3.5
17.11. Sparse EIF SLAM - Part 1 08.pdf MP4 Sheet5 Octave-Code Probabilistic Robotics Book, Chapter 12.1-12.7 SEIF Paper
23.11. Sparse EIF SLAM - Part 2 MP4 Sheet6 Octave-Code Probabilistic Robotics Book, Chapter 12.1-12.7 SEIF Paper SEIF: Insights on Sparsification
23.11. Summary on the Kalman Filter and its Friends for SLAM 09.pdf ---
30.11. Grid Maps 10.pdf MP4 --- Probabilistic Robotics Book, Chapter 4.2, 9.1-9.2
30.11. Short Introduction to the Particle Filter and PF Localization 11.pdf Sheet7 Octave-Code-1 Octave-Code-2 Probabilistic Robotics Book, Chapter 8.3
03.12. FastSLAM 12.pdf MP4 Sheet8 Octave-Code Probabilistic Robotics Book, Chapter 13.1-13.3 + 13.8 FastSLAM: A Factored Solution to the Simultaneous Localization and Mapping Problem Simultaneous Localization and Mapping with Unknown Data Association Using FastSLAM
10.12. Grid-Based SLAM with Rao-Blackwellized Particle Filters 13.pdf MP4 --- Improved Techniques for Grid Mapping with Rao- Blackwellized Particle Filters Analyzing Gaussian Proposal Distributions for Mapping with Rao-Blackwellized Particle Filters Probabilistic Robotics Book, Chapter 13.10
14.12. Least-Squares 14.pdf MP4 Sheet9 Octave-Code Methods for Non-Linear Least Squares Probelms Wiki:Gauss-Newton Probabilistic Robotics Book, Chapter 11.4
18.01. Least-Squares Approach to SLAM 15.pdf MP4 Sheet10 Octave-Code A Tutorial on Graph-Based SLAM Probabilistic Robotics Book, Chapter 11
25.01. Hierarchical Pose-Graphs for Online Mapping 16.pdf MP4 --- Hierarchical Optimization on Manifolds for Online 2D and 3D Mapping
25.01. Graph-Based SLAM with Landmarks 17.pdf --- Probabilistic Robotics Book, Chapter 11 Methods for Non-Linear Least Squares Probelms
01.02. Max-Mixture and Robust Least Squares for SLAM 18.pdf MP4 --- Inference on Networks of Mixtures for Robust Robot Mapping Methods for Non-Linear Least Squares Problems
01.02. Stochastic Gradient Descent for SLAM 19.pdf --- Non-linear Constraint Network Optimization for Efficient Map Learning Fast Iterative Optimization of Pose Graphs with Poor Initial Estimates
08.02. Front-Ends for Graph-Based SLAM 20.pdf MP4 --- A method for Registration of 3-D Shapes Real-Time Correlative Scan Matching FLIRT -- Interest Regions for 2D Range Data Recognizing Places using Spectrally Clustered Local Matches
08.02. Short Summary 21.pdf --- ---

In case you need to revisit material about Gaussians or a reference for matrix operations:

Relevant Literature for the Course

Most of the literature is available as PDF files for free, but not the "Probabilistic Robotics" book. You find it in the TF library.