Topological mapping using spectral clustering and classification
- 1 October 2007
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3491-3496
- https://doi.org/10.1109/iros.2007.4399611
Abstract
In this work we present an online method for generating topological maps from raw sensor information. We first describe an algorithm to automatically decompose a map into submap segments using a graph partitioning technique known as spectral clustering. We then describe how to train a classifier to recognize graph submaps from laser signatures using the AdaBoost machine learning algorithm. We demonstrate that the we can perform topological mapping by incrementally segmenting the world as the robot moves through its environment, and we can close the loop when the learned classifier recognizes that the robot has returned to a previously visited location.Keywords
This publication has 6 references indexed in Scilit:
- Outdoor SLAM using visual appearance and laser rangingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2006
- Using the topological skeleton for scalable global metrical map-buildingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Simultaneous Localization and Map Building in Large-Scale Cyclic Environments Using the Atlas FrameworkThe International Journal of Robotics Research, 2004
- Local metrical and global topological maps in the hybrid spatial semantic hierarchyPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- Simultaneous map building and localization for an autonomous mobile robotPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Learning metric-topological maps for indoor mobile robot navigationArtificial Intelligence, 1998