Transcription of A Tutorial on Graph-Based SLAM - uni-freiburg.de
{{id}} {{{paragraph}}}
Example: bankruptcy
A Tutorial on Graph-Based SLAM - uni-freiburg.de
1A Tutorial on Graph-Based SLAMG iorgio GrisettiRainer K ummerleCyrill StachnissWolfram BurgardDepartment of Computer Science, University of Freiburg, 79110 Freiburg, GermanyAbstract Being able to build a map of the environment andto simultaneously localize within this map is an essential skill formobile robots navigating in unknown environments in absenceof external referencing systems such as GPS. This so-calledsimultaneous localization and mapping ( slam ) problem hasbeen one of the most popular research topics in mobile roboticsfor the last two decades and efficient approaches for solving thistask have been proposed. One intuitive way of formulating SLAMis to use a graph whose nodes correspond to the poses of the robotat different points in time and whose edges represent constraintsbetween the poses. The latter are obtained from observationsof the environment or from movement actions carried out bythe robot. Once such a graph is constructed, the map can becomputed by finding the spatial configuration of the nodes thatis mostly consistent with the measurements modeled by theedges.
Consequently, graph-based SLAM methods have undergone a renaissance and currently belong to the state-of-the-art techniques with respect to speed and accuracy. The aim of this tutorial is to introduce the SLAM problem in its probabilistic form and to guide the reader to the synthesis of an effective and state-of-the-art graph-based SLAM method.
Loading..
Tags:
Information
Domain:
Source:
Link to this page:
Please notify us if you found a problem with this document: