{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,14]],"date-time":"2026-01-14T17:58:54Z","timestamp":1768413534499,"version":"3.49.0"},"reference-count":32,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2024,2,9]],"date-time":"2024-02-09T00:00:00Z","timestamp":1707436800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>An accurate inverse dynamics model of manipulators can be effectively learned using neural networks. However, further research is required to investigate the impact of spatiotemporal variations in manipulator motion sequences on network learning. In this work, the Velocity Aware Spatial-Temporal Attention Residual LSTM neural network (VA-STA-ResLSTM) is proposed to learn a more accurate inverse dynamics model, which uses a velocity-aware spatial-temporal attention mechanism to extract dynamic spatiotemporal features selectively from the motion sequence of the serial manipulator.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>The multi-layer perception (MLP) attention mechanism is adopted to capture the correlation between joint position and velocity in the motion sequence, and the state correlation between hidden units in the LSTM network to reduce the weight of invalid features. A velocity-aware state fusion approach of LSTM network hidden units' states is proposed, which utilizes variation in joint velocity to adapt to the temporal characteristics of the manipulator dynamic motion, improving the generalization and accuracy of the neural network.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Comparative experiments have been conducted on two open datasets and a self-built dataset. Specifically, the proposed method achieved an average accuracy improvement of 61.88% and 43.93% on the two different open datasets and 71.13% on the self-built dataset compared to the LSTM network. These results demonstrate a significant advancement in accuracy for the proposed method.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>Compared with the state-of-the-art inverse dynamics model learning methods of manipulators, the modeling accuracy of the proposed method in this paper is higher by an average of 10%. Finally, by visualizing attention weights to explain the training procedure, it was found that dynamic modeling only relies on partial features, which is meaningful for future optimization of inverse dynamic model learning methods.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fnbot.2024.1353879","type":"journal-article","created":{"date-parts":[[2024,2,9]],"date-time":"2024-02-09T04:38:00Z","timestamp":1707453480000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":6,"title":["Velocity-aware spatial-temporal attention LSTM model for inverse dynamic model learning of manipulators"],"prefix":"10.3389","volume":"18","author":[{"given":"Wenhui","family":"Huang","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yunhan","family":"Lin","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mingxin","family":"Liu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Huasong","family":"Min","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2024,2,9]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1729881420908076","DOI":"10.1177\/1729881420908076","article-title":"Path planning optimization of six-degree-of-freedom robotic manipulators using evolutionary algorithms","volume":"17","author":"Baressi \u0160egota","year":"2020","journal-title":"International J. Adv. Robot. Syst."},{"key":"B2","doi-asserted-by":"publisher","first-page":"1061","DOI":"10.1109\/LRA.2022.3222951","article-title":"Hybrid learning of time-series inverse dynamics models for locally isotropic robot motion","volume":"8","author":"\u00c7allar","year":"2022","journal-title":"IEEE Robot. Autom. Lett."},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/W14-4012","article-title":"On the properties of neural machine translation: Encoder-decoder approaches","author":"Cho","year":"2014","journal-title":"arXiv preprint arXiv:1409.1259"},{"key":"B4","doi-asserted-by":"publisher","first-page":"348","DOI":"10.1016\/j.neucom.2020.04.110","article-title":"Interpretable spatio-temporal attention LSTM model for flood forecasting","volume":"403","author":"Ding","year":"2020","journal-title":"Neurocomputing"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01270-0_23","article-title":"\u201cInteraction-aware spatio-temporal pyramid attention networks for action classification,\u201d","author":"Du","year":"2018","journal-title":"Proceedings of the European Conference on Computer Vision (ECCV)"},{"key":"B6","doi-asserted-by":"crossref","DOI":"10.1109\/IROS.2013.6697198","article-title":"\u201cIdentification of standard dynamic parameters of robots with positive definite inertia matrix,\u201d","volume-title":"2013 IEEE\/RSJ International Conference on Intelligent Robots and Systems","author":"Gautier","year":"2013"},{"key":"B7","doi-asserted-by":"publisher","first-page":"e202100009","DOI":"10.1002\/gamm.202100009","article-title":"Structured learning of rigid-body dynamics: a survey and unified view from a robotics perspective","volume":"44","author":"Geist","year":"2021","journal-title":"GAMM-Mitteilungen"},{"key":"B8","doi-asserted-by":"publisher","first-page":"2222","DOI":"10.1109\/TNNLS.2016.2582924","article-title":"LSTM: a search space odyssey","volume":"28","author":"Greff","year":"2016","journal-title":"IEEE Trans. Neur. Netw. Learn. Syst."},{"key":"B9","doi-asserted-by":"publisher","first-page":"9590","DOI":"10.1109\/TITS.2022.3155613","article-title":"A dynamic spatial-temporal attention network for early anticipation of traffic accidents","volume":"23","author":"Karim","year":"2022","journal-title":"IEEE Trans. Intell. Transport. Syst."},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2018\/324","article-title":"\u201cHST-LSTM: A hierarchical spatial-temporal long-short term memory network for location prediction,\u201d","author":"Kong","year":"2018","journal-title":"IJCAI"},{"key":"B11","doi-asserted-by":"publisher","first-page":"88644","DOI":"10.1109\/ACCESS.2022.3192430","article-title":"A combined model for multivariate time series forecasting based on MLP-feedforward attention-LSTM","volume":"10","author":"Liu","year":"2022","journal-title":"IEEE Access"},{"key":"B12","doi-asserted-by":"publisher","first-page":"102360","DOI":"10.1016\/j.rcim.2022.102360","article-title":"Robot learning towards smart robotic manufacturing: a review","volume":"77","author":"Liu","year":"2022","journal-title":"Robot. Comput. Integr. Manufact."},{"key":"B13","doi-asserted-by":"publisher","first-page":"1307","DOI":"10.1007\/s40997-023-00596-3","article-title":"Modeling and control of robotic manipulators based on artificial neural networks: a review","volume":"47","author":"Liu","year":"2023","journal-title":"Iranian J. Sci. Technol. Trans. Mechan. Eng."},{"key":"B14","article-title":"\u201cIncremental local gaussian regression,\u201d","author":"Meier","year":"2014","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B15","doi-asserted-by":"crossref","first-page":"2251","DOI":"10.1109\/TENCON.2019.8929622","article-title":"\u201cModel learning for robotic manipulators using recurrent neural networks,\u201d","volume-title":"TENCON 2019\u20132019 IEEE Region 10 Conference (TENCON)","author":"Mukhopadhyay","year":"2019"},{"key":"B16","doi-asserted-by":"publisher","first-page":"2015","DOI":"10.1163\/016918609X12529286896877","article-title":"Model learning with local gaussian process regression","volume":"23","author":"Nguyen-Tuong","year":"2009","journal-title":"Adv. Robot."},{"key":"B17","doi-asserted-by":"publisher","first-page":"281","DOI":"10.1016\/j.neucom.2018.11.066","article-title":"Machine learning on sequential data using a recurrent weighted average","volume":"331","author":"Ostmeyer","year":"2019","journal-title":"Neurocomputing"},{"key":"B18","doi-asserted-by":"crossref","first-page":"3442","DOI":"10.1109\/IROS.2015.7353857","article-title":"\u201cReal-time deep learning of robotic manipulator inverse dynamics,\u201d","volume-title":"2015 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)","author":"Polydoros","year":"2015"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2022.XVIII.066","article-title":"End-to-end learning of hybrid inverse dynamics models for precise and compliant impedance control","author":"Reuss","year":"2022","journal-title":"arXiv preprint arXiv:2205.13804"},{"key":"B20","doi-asserted-by":"crossref","first-page":"811","DOI":"10.1109\/HUMANOIDS.2017.8246965","article-title":"\u201cLearning inverse dynamics models in o (n) time with lstm networks,\u201d","volume-title":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","author":"Rueckert","year":"2017"},{"key":"B21","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1142\/S0129065704001899","article-title":"Gaussian processes for machine learning","volume":"14","author":"Seeger","year":"2004","journal-title":"Int. J. Neural Syst."},{"key":"B22","doi-asserted-by":"publisher","first-page":"110461","DOI":"10.1016\/j.asoc.2023.110461","article-title":"Residual LSTM based short-term load forecasting","volume":"144","author":"Sheng","year":"2023","journal-title":"Appl. Soft Comput."},{"key":"B23","doi-asserted-by":"publisher","first-page":"3459","DOI":"10.1109\/TIP.2018.2818328","article-title":"Spatio-temporal attention-based LSTM networks for 3D action recognition and detection","volume":"27","author":"Song","year":"2018","journal-title":"IEEE Trans. Image Proc."},{"key":"B24","doi-asserted-by":"publisher","first-page":"2133","DOI":"10.1080\/01431160802549278","article-title":"How many hidden layers and nodes?","volume":"30","author":"Stathakis","year":"2009","journal-title":"Int. J. Remote Sens."},{"key":"B25","doi-asserted-by":"crossref","first-page":"2906","DOI":"10.1109\/ICASSP.2018.8461750","article-title":"\u201cAdvanced LSTM: a study about better time dependency modeling in emotion recognition,\u201d","volume-title":"2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","author":"Tao","year":"2018"},{"key":"B26","doi-asserted-by":"crossref","first-page":"14","DOI":"10.1007\/978-3-031-24801-6_2","article-title":"\u201cInnovative soft computing techniques for the evaluation of the mechanical stress state of steel plates,\u201d","volume-title":"International Conference on Applied Intelligence and Informatics","author":"Versaci","year":"2022"},{"key":"B27","unstructured":"VijayakumarS.\n            D'SouzaA.\n            SchaalS.\n          16212764LWPR: A Scalable Method for Incremental Online Learning in High Dimensions2005"},{"key":"B28","article-title":"\u201cGaussian processes for regression,\u201d","author":"Williams","year":"1995","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2021\/225","article-title":"Spatial-temporal sequential hypergraph network for crime prediction with dynamic multiplex relation learning","author":"Xia","year":"2022","journal-title":"arXiv preprint arXiv:2201.02435"},{"key":"B30","doi-asserted-by":"publisher","first-page":"1675","DOI":"10.1109\/TASLP.2019.2925934","article-title":"Speech emotion classification using attention-based LSTM","volume":"27","author":"Xie","year":"2019","journal-title":"IEEE\/ACM Trans. Audio, Speech, Lang. Proc."},{"key":"B31","doi-asserted-by":"crossref","first-page":"1387","DOI":"10.1109\/ICRA40945.2020.9197445","article-title":"\u201cNeural network based inverse dynamics identification and external force estimation on the da Vinci Research Kit,\u201d","volume-title":"2020 IEEE International Conference on Robotics and Automation (ICRA)","author":"Yilmaz","year":"2020"},{"key":"B32","article-title":"\u201cWhen can self-attention be replaced by feed forward layers?\u201d","author":"Zhang","year":"2020","journal-title":"arXiv preprint arXiv:2005.13895"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2024.1353879\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,2,9]],"date-time":"2024-02-09T04:38:08Z","timestamp":1707453488000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2024.1353879\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,2,9]]},"references-count":32,"alternative-id":["10.3389\/fnbot.2024.1353879"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2024.1353879","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,2,9]]},"article-number":"1353879"}}