Deep learning for 2D scan matching and loop closure

Simple item page
dc.contributor.authorLi, J.
dc.contributor.authorZhan, H.
dc.contributor.authorChen, B.
dc.contributor.authorReid, I.
dc.contributor.authorLee, G.
dc.contributor.conferenceIEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (24 Sep 2017 - 28 Sep 2017 : Vancouver, Canada)
dc.contributor.editorBicchi, A.
dc.contributor.editorOkamura, A.
dc.date.issued2017
dc.description.abstractAlthough 2D LiDAR based Simultaneous Localization and Mapping (SLAM) is a relatively mature topic nowadays, the loop closure problem remains challenging due to the lack of distinctive features in 2D LiDAR range scans. Existing research can be roughly divided into correlation based approaches e.g. scan-to-submap matching and feature based methods e.g. bag-of-words (BoW). In this paper, we solve loop closure detection and relative pose transformation using 2D LiDAR within an end-to-end Deep Learning framework. The algorithm is verified with simulation data and on an Unmanned Aerial Vehicle (UAV) flying in indoor environment. The loop detection ConvNet alone achieves an accuracy of 98.2% in loop closure detection. With a verification step using the scan matching ConvNet, the false positive rate drops to around 0.001%. The proposed approach processes 6000 pairs of raw LiDAR scans per second on a Nvidia GTX1080 GPU.
dc.description.statementofresponsibilityJiaxin Li, Huangying Zhan, Ben M. Chen, Ian Reid, Gim Hee Lee
dc.identifier.citationProceedings of the ... IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE/RSJ International Conference on Intelligent Robots and Systems, 2017 / Bicchi, A., Okamura, A. (ed./s), vol.2017-September, pp.763-768
dc.identifier.doi10.1109/IROS.2017.8202236
dc.identifier.isbn9781538626825
dc.identifier.issn2153-0858
dc.identifier.issn2153-0866
dc.identifier.orcidReid, I. [0000-0001-7790-6423]
dc.identifier.urihttp://hdl.handle.net/2440/117299
dc.language.isoen
dc.publisherIEEE
dc.relation.ispartofseriesIEEE International Conference on Intelligent Robots and Systems
dc.rights© 2017 IEEE
dc.source.urihttps://doi.org/10.1109/iros.2017.8202236
dc.titleDeep learning for 2D scan matching and loop closure
dc.typeConference paper
pubs.publication-statusPublished

Files

Collections

Computer Science publications
Computer Vision publications