Detailansicht

Inhaltsangabe1 Introduction 1.1 Applications of SLAM 1.2 Joint Estimation 1.3 Posterior Estimation 1.4 The Extended Kalman Filter 1.5 Structure and Sparsity in SLAM 1.6 FastSLAM 1.7 Outline 2 The SLAM Problem 2.1 Problem Definition 2.2 SLAM Posterior 2.3 SLAM as a Markov Chain 2.4 Extended Kalman Filtering 2.5 Scaling SLAM Algorithms 2.6 Robust Data Association 2.7 Comparison of FastSLAM to Existing Techniques 3 FastSLAM 1.0 3.1 Particle Filtering 3.2 Factored Posterior Representation 3.3 The FastSLAM 1.0 Algorithm 3.4 FastSLAM with Unknown Data Association 3.5 Summary of the FastSLAM Algorithm 3.6 FastSLAM Extensions 3.7 Log(N) FastSLAM 3.8 Experimental Results 3.9 Summary 4 FastSLAM 2.0 4.1 Sample Impoverishment 4.2 FastSLAM 2.0 4.3 FastSLAM 2.0 Convergence 4.4 Experimental Results 4.5 Grid-based FastSLAM 4.6 Summary 5 Dynamic Environments 5.1 SLAM With Dynamic Landmarks 5.2 Simultaneous Localization and People Tracking 5.3 FastSLAP Implementation 5.4 Experimental Results 5.5 Summary 6 Conclusions 6.1 Conclusions 6.2 Future Work References Index

This monograph describes a new family of algorithms for the simultaneous localization and mapping problem in robotics (SLAM). SLAM addresses the problem of acquiring an environment map with a roving robot, while simultaneously localizing the robot relative to this map. This problem has received enormous attention in the robotics community in the past few years, reaching a peak of popularity on the occasion of the DARPA Grand Challenge in October 2005, which was won by the team headed by the authors. The FastSLAM family of algorithms applies particle filters to the SLAM Problem, which provides new insights into the data association problem that is paramount in SLAM. The FastSLAM-type algorithms have enabled robots to acquire maps of unprecedented size and accuracy, in a number of robot application domains and have been successfully applied in different dynamic environments, including the solution to the problem of people tracking.