A digital twin of intelligent robotic grasping based on single-loop-optimized differentiable architecture search and sim-real collaborative learning

Simple item page
dc.check.date2025-10-14en
dc.check.infoAccess to this article is restricted until 12 months after publication by request of the publisheren
dc.contributor.authorJiao, Qingen
dc.contributor.authorHu, Weifeien
dc.contributor.authorHao, Guangboen
dc.contributor.authorCheng, Jinen
dc.contributor.authorPeng, Xiangen
dc.contributor.authorLiu, Zhenyuen
dc.contributor.authorTan, Jianrongen
dc.contributor.funderNational Natural Science Foundation of Chinaen
dc.contributor.funderKey Research and Development Program of Zhejiang Provinceen
dc.contributor.funderNatural Science Foundation of Zhejiang Provinceen
dc.date.accessioned2024-11-25T11:57:22Z
dc.date.available2024-11-25T11:57:22Z
dc.date.issued2024-10-14en
dc.description.abstractThe effectiveness of deep learning models for vision-based intelligent robotic grasping (IRG) tasks typically hinges upon the deep neural network (DNN) architecture as well as the task-oriented annotated training samples. Nevertheless, current methods applied for designing DNN architectures depend on human expertise or discrete search by evolution and reinforcement learning algorithms, which leads to enormous computational cost. Moreover, DNNs trained solely on simulation-labeled data face challenges in direct real-world deployment. In response to these concerns, this paper proposes a new stable and fast differentiable architecture search method (DARTS) based on a single-loop optimization framework, named single-loop-optimized DARTS (SLO-DARTS). This method enables simultaneous updates to the weights and architecture parameters of neural networks by continuously relaxing the discrete search space. Additionally, a digital twin (DT) framework integrating the Grasp-CycleGAN method is developed to minimize the visual gap between simulated and real-world IRG scenarios, enhancing the transferability of DNNs trained in simulation. The DT framework can not only enhance the IRG accuracy but also save the costly expense of large-scale real labeled data collection. Experiments demonstrate that the proposed SLO-DARTS method achieves a time-efficient optimization process while delivering a DNN with improved prediction accuracy compared to the original dual-loop-optimized DARTS method. The developed DT framework produces IRG accuracies of 92.6%, 86.3%, and 83.7% for single household objects, single adversarial objects, and cluttered objects, respectively.en
dc.description.sponsorshipNational Natural Science Foundation of China (Grant numbers 52275275; U22A6001); Natural Science Foundation of Zhejiang Province (Grant Number LZ22E050006); Key Research and Development Program of Zhejiang Province (Grant Number 2023C01008).en
dc.description.statusPeer revieweden
dc.description.versionAccepted Versionen
dc.format.mimetypeapplication/pdfen
dc.identifier.citationJiao, Q., Hu, W., Hao, G., Cheng, J., Peng, X., Liu, Z. and Tan, J. (2024) 'A digital twin of intelligent robotic grasping based on single-loop-optimized differentiable architecture search and sim-real collaborative learning', Journal of Intelligent Manufacturing. https://doi.org/10.1007/s10845-024-02498-wen
dc.identifier.doihttps://doi.org/10.1007/s10845-024-02498-wen
dc.identifier.eissn1572-8145en
dc.identifier.endpage20en
dc.identifier.issn0956-5515en
dc.identifier.journaltitleJournal of Intelligent Manufacturingen
dc.identifier.startpage1en
dc.identifier.urihttps://hdl.handle.net/10468/16668
dc.language.isoenen
dc.publisherSpringer Natureen
dc.rights© 2024, the Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024. This is a post-peer-review, pre-copyedit version of an article published in the Journal of Intelligent Manufacturing. The final authenticated version is available online at: https://doi.org/10.1007/s10845-024-02498-wen
dc.subjectRobotic graspingen
dc.subjectDeep neural networken
dc.subjectArchitecture searchen
dc.subjectCollaborative learningen
dc.subjectDigital twinen
dc.subjectAutomatic designen
dc.titleA digital twin of intelligent robotic grasping based on single-loop-optimized differentiable architecture search and sim-real collaborative learningen
dc.typeArticle (peer-reviewed)en
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Manuscript_robotic_grasp_final_version (accepted).pdf
Size:
1.97 MB
Format:
Adobe Portable Document Format
Description:
Accepted Version
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
2.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Electrical and Electronic Engineering - Journal Articles