Hide metadata

dc.date.accessioned2019-04-23T10:04:50Z
dc.date.available2019-04-23T10:04:50Z
dc.date.created2019-01-22T15:38:03Z
dc.date.issued2018
dc.identifier.citationMiseikis, Justinas Brijacak, Inka Yahyanejad, Saeed Glette, Kyrre Elle, Ole Jacob Tørresen, Jim . Transfer Learning for Unseen Robot Detection and Joint Estimation on a Multi-Objective Convolutional Neural Network. 2018 IEEE International Conference on Intelligence and Safety for Robotics (ISR 2018). 2018, 337-342 IEEE
dc.identifier.urihttp://hdl.handle.net/10852/67772
dc.description.abstractA significant problem of using deep learning techniques is the limited amount of data available for training. There are some datasets available for the popular problems like item recognition and classification or self-driving cars, however, it is very limited for the industrial robotics field. In previous work, we have trained a multi-objective Convolutional Neural Network (CNN) to identify the robot body in the image and estimate 3D positions of the joints by using just a 2D image, but it was limited to a range of robots produced by Universal Robots (UR). In this work, we extend our method to work with a new robot arm - Kuka LBR iiwa, which has a significantly different appearance and an additional joint. However, instead of collecting large datasets once again, we collect a number of smaller datasets containing a few hundred frames each and use transfer learning techniques on the CNN trained on UR robots to adapt it to a new robot having different shapes and visual features. We have proven that transfer learning is not only applicable in this field, but it requires smaller well-prepared training datasets, trains significantly faster and reaches similar accuracy compared to the original method, even improving it on some aspects.en_US
dc.languageEN
dc.publisherIEEE
dc.relation.ispartofMišeikis, Justinas (2019) An Environment-Aware Robot Arm Platform Using Low-Cost Sensors and Deep Learning. Doctoral thesis. http://hdl.handle.net/10852/70397
dc.relation.urihttp://hdl.handle.net/10852/70397
dc.titleTransfer Learning for Unseen Robot Detection and Joint Estimation on a Multi-Objective Convolutional Neural Networken_US
dc.typeChapteren_US
dc.creator.authorMiseikis, Justinas
dc.creator.authorBrijacak, Inka
dc.creator.authorYahyanejad, Saeed
dc.creator.authorGlette, Kyrre
dc.creator.authorElle, Ole Jacob
dc.creator.authorTørresen, Jim
cristin.unitcode185,15,5,42
cristin.unitnameForskningsgruppe for robotikk og intelligente systemer
cristin.ispublishedtrue
cristin.fulltextpostprint
dc.identifier.cristin1663192
dc.identifier.bibliographiccitationinfo:ofi/fmt:kev:mtx:ctx&ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.btitle=2018 IEEE International Conference on Intelligence and Safety for Robotics (ISR 2018)&rft.spage=337&rft.date=2018
dc.identifier.startpage337
dc.identifier.endpage342
dc.identifier.pagecount613
dc.identifier.doihttps://doi.org/10.1109/IISR.2018.8535937
dc.identifier.urnURN:NBN:no-70937
dc.type.documentBokkapittelen_US
dc.type.peerreviewedPeer reviewed
dc.source.isbn9781538655481
dc.identifier.fulltextFulltext https://www.duo.uio.no/bitstream/handle/10852/67772/2/miseikis-isr2018.pdf
dc.type.versionAcceptedVersion
cristin.btitle2018 IEEE International Conference on Intelligence and Safety for Robotics (ISR 2018)
dc.relation.projectNFR/240862


Files in this item

Appears in the following Collection

Hide metadata