dc.contributor.author | Dyrstad, Jonatan Sjølund | |
dc.contributor.author | Mathiassen, John Reidar Bartle | |
dc.date.accessioned | 2018-04-26T12:25:22Z | |
dc.date.available | 2018-04-26T12:25:22Z | |
dc.date.created | 2018-03-13T09:26:48Z | |
dc.date.issued | 2018-03-26 | |
dc.identifier.citation | Robotics and Biomimetics (ROBIO), 2017 IEEE International Conference on | nb_NO |
dc.identifier.issn | 2197-3768 | |
dc.identifier.uri | http://hdl.handle.net/11250/2496162 | |
dc.description.abstract | We present an approach to robotic deep learning from demonstration in virtual reality, which combines a deep 3D convolutional neural network, for grasp detection from 3D point clouds, with domain randomization to generate a large training data set. The use of virtual reality (VR) enables robot learning from demonstration in a virtual environment. In this environment, a human user can easily and intuitively demonstrate examples of how to grasp an object, such as a fish. From a few dozen of these demonstrations, we use domain randomization to generate a large synthetic training data set consisting of 76 000 example grasps of fish. After training the network using this data set, the network is able to guide a gripper to grasp virtual fish with good success rates. Our domain randomization approach is a step towards an efficient way to perform robotic deep learning from demonstration in virtual reality. | nb_NO |
dc.language.iso | eng | nb_NO |
dc.publisher | |IEEE Xplore Digital Library | 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 | Grasping virtual fish: A step towards deep learning from demonstration in virtual reality | nb_NO |
dc.type | Journal article | nb_NO |
dc.type | Peer reviewed | nb_NO |
dc.description.version | publishedVersion | nb_NO |
dc.rights.holder | The authors | nb_NO |
dc.source.journal | Robotics and Biomimetics | nb_NO |
dc.identifier.doi | 10.1109/ROBIO.2017.8324578 | |
dc.identifier.cristin | 1572396 | |
cristin.unitcode | 7566,5,0,0 | |
cristin.unitname | Prosessteknologi | |
cristin.ispublished | false | |
cristin.fulltext | original | |
cristin.qualitycode | 1 | |