dc.contributor.author | Dyrstad, Jonatan Sjølund | |
dc.contributor.author | Bakken, Marianne | |
dc.contributor.author | Grøtli, Esten Ingar | |
dc.contributor.author | Schulerud, Helene | |
dc.contributor.author | Mathiassen, John Reidar Bartle | |
dc.date.accessioned | 2019-07-08T12:25:28Z | |
dc.date.available | 2019-07-08T12:25:28Z | |
dc.date.created | 2019-01-25T12:51:11Z | |
dc.date.issued | 2018-12-12 | |
dc.identifier.citation | 2018 IEEE International Conference on Robotics and Biomimetrics (ROBIO) | nb_NO |
dc.identifier.issn | 2197-3768 | |
dc.identifier.uri | http://hdl.handle.net/11250/2603749 | |
dc.description.abstract | We consider the case of robotic bin picking of reflective steel parts, using a structured light 3D camera as a depth imaging device. In this paper, we present a new method for bin picking, based on a dual-resolution convolutional neural network trained entirely in a simulated environment. The dualresolution network consists of a high resolution focus network to compute the grasp and a low resolution context network to avoid local collisions.The reflectivity of the steel parts result in depth images that have a lot of missing data. To take this into account, training of the neural net is done by domain randomization on a large set of synthetic depth images that simulate the missing data problems of the real depth images. We demonstrate both in simulation and in a real-world test that our method can perform bin picking of reflective steel parts | nb_NO |
dc.description.sponsorship | This project was funded by the Research Council of Norway through the SFI Manufacturing Centre for Researchbased Innovation, and in part from Research Council of Norway Grant No. 262900. | nb_NO |
dc.language.iso | eng | nb_NO |
dc.publisher | Institute of Electrical and Electronics Engineers (IEEE) | nb_NO |
dc.rights | Attribution-NonCommercial-NoDerivatives 4.0 Internasjonal | * |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/4.0/deed.no | * |
dc.title | Bin Picking of Reflective Steel Parts using a Dual-Resolution Convolutional Neural Network Trained in a Simulated Environment | nb_NO |
dc.type | Journal article | nb_NO |
dc.type | Peer reviewed | nb_NO |
dc.description.version | acceptedVersion | nb_NO |
dc.source.journal | Robotics and Biomimetics | nb_NO |
dc.identifier.doi | 10.1109/ROBIO.2018.8664766 | |
dc.identifier.cristin | 1665104 | |
dc.relation.project | Norges forskningsråd: 262900 | nb_NO |
cristin.unitcode | 7401,90,26,0 | |
cristin.unitcode | 7401,90,33,0 | |
cristin.unitcode | 7566,2,0,0 | |
cristin.unitname | Mathematics and Cybernetics | |
cristin.unitname | Smart Sensor Systems | |
cristin.unitname | Sjømatteknologi | |
cristin.ispublished | false | |
cristin.fulltext | postprint | |
cristin.qualitycode | 1 | |