{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,19]],"date-time":"2025-11-19T19:20:06Z","timestamp":1763580006680,"version":"3.45.0"},"reference-count":53,"publisher":"American Association for the Advancement of Science (AAAS)","issue":"108","funder":[{"DOI":"10.13039\/100023581","name":"National Science Foundation Graduate Research Fellowship Program","doi-asserted-by":"publisher","award":["DGE-2039655"],"award-info":[{"award-number":["DGE-2039655"]}],"id":[{"id":"10.13039\/100023581","id-type":"DOI","asserted-by":"publisher"}]},{"name":"National Science Foundation Foundational Research in Robotics","award":["2233164"],"award-info":[{"award-number":["2233164"]}]},{"name":"National Science Foundation Foundational Research in Robotics","award":["2328051"],"award-info":[{"award-number":["2328051"]}]},{"name":"National Science Foundation Foundational Research in Robotics","award":["2328050"],"award-info":[{"award-number":["2328050"]}]}],"content-domain":{"domain":["www.science.org"],"crossmark-restriction":true},"short-container-title":["Sci. Robot."],"published-print":{"date-parts":[[2025,11,19]]},"abstract":"<jats:p>\n                    Data-driven methods have transformed our ability to assess and respond to human movement with wearable robots, promising real-world rehabilitation and augmentation benefits. However, the proliferation of data-driven methods, with the associated demand for increased personalization and performance, requires vast quantities of high-quality, device-specific data. Procuring these data is often intractable because of resource and personnel costs. We propose a framework that overcomes data scarcity by leveraging simulated sensors from biomechanical models to form a stepping-stone domain through which easily accessible data can be translated into data-limited domains. We developed and optimized a deep domain adaptation network that replaces costly, device-specific, labeled data with open-source datasets and unlabeled exoskeleton data. Using our network, we trained a hip and knee joint moment estimator with performance comparable to a best-case model trained with a complete, device-specific dataset [incurring only an 11 to 20%, 0.019 to 0.028 newton-meters per kilogram (Nm\/kg) increase in error for a semisupervised model and 20 to 44%, 0.033 to 0.062 Nm\/kg for an unsupervised model]. Our network significantly outperformed counterpart networks without domain adaptation (which incurred errors of 36 to 45% semisupervised and 50 to 60% unsupervised). Deploying our models in the real-time control loop of a hip\/knee exoskeleton (\n                    <jats:italic toggle=\"yes\">N<\/jats:italic>\n                    \u00a0=\u00a08) demonstrated estimator performance similar to offline results while augmenting user performance based on those estimated moments (9.5 to 14.6% metabolic cost reductions compared with no exoskeleton). Our framework enables researchers to train real-time deployable deep learning, task-agnostic models with limited or no access to labeled, device-specific data.\n                  <\/jats:p>","DOI":"10.1126\/scirobotics.ads8652","type":"journal-article","created":{"date-parts":[[2025,11,19]],"date-time":"2025-11-19T18:58:23Z","timestamp":1763578703000},"update-policy":"https:\/\/doi.org\/10.34133\/aaas_crossmark","source":"Crossref","is-referenced-by-count":0,"title":["Deep domain adaptation eliminates costly data required for task-agnostic wearable robotic control"],"prefix":"10.1126","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9496-8079","authenticated-orcid":true,"given":"Keaton L.","family":"Scherpereel","sequence":"first","affiliation":[{"name":"George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332, USA."},{"name":"Institute for Robotics and Intelligent Machines, Georgia Institute of Technology, Atlanta, GA 30332, USA."}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5321-6038","authenticated-orcid":true,"given":"Matthew C.","family":"Gombolay","sequence":"additional","affiliation":[{"name":"Institute for Robotics and Intelligent Machines, Georgia Institute of Technology, Atlanta, GA 30332, USA."},{"name":"School of Interactive Computing, Georgia Institute of Technology, Atlanta, GA 30332, USA."}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6066-9222","authenticated-orcid":true,"given":"Max K.","family":"Shepherd","sequence":"additional","affiliation":[{"name":"College of Engineering, Bouv\u00e9 College of Health Sciences, and Institute for Experiential Robotics, Northeastern University, Boston, MA 02115, USA."}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6774-7254","authenticated-orcid":true,"given":"Carlos A.","family":"Carrasquillo","sequence":"additional","affiliation":[{"name":"George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332, USA."},{"name":"Institute for Robotics and Intelligent Machines, Georgia Institute of Technology, Atlanta, GA 30332, USA."}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7952-1794","authenticated-orcid":true,"given":"Omer T.","family":"Inan","sequence":"additional","affiliation":[{"name":"School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA 30332, USA."}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5376-2258","authenticated-orcid":true,"given":"Aaron J.","family":"Young","sequence":"additional","affiliation":[{"name":"George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332, USA."},{"name":"Institute for Robotics and Intelligent Machines, Georgia Institute of Technology, Atlanta, GA 30332, USA."}]}],"member":"221","reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2022.3191039"},{"key":"e_1_3_2_3_2","doi-asserted-by":"crossref","unstructured":"I. Kang P. Kunapuli H. Hsu A. J. Young \u201cElectromyography (EMG) signal contributions in speed and slope estimation using robotic exoskeletons \u201d in 2019 IEEE 16th International Conference on Rehabilitation Robotics (ICORR) (IEEE 2019) pp. 548\u2013553.","DOI":"10.1109\/ICORR.2019.8779433"},{"key":"e_1_3_2_4_2","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2022.3147565"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.1109\/TMRB.2019.2961749"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2021.3062562"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.1007\/s42235-022-00230-z"},{"key":"e_1_3_2_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2022.3173426"},{"key":"e_1_3_2_9_2","doi-asserted-by":"publisher","DOI":"10.3390\/s23125404"},{"key":"e_1_3_2_10_2","doi-asserted-by":"publisher","DOI":"10.3390\/s21082807"},{"key":"e_1_3_2_11_2","doi-asserted-by":"publisher","DOI":"10.1109\/OJCSYS.2022.3165733"},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1126\/scirobotics.adi8852"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-024-08157-7"},{"key":"e_1_3_2_14_2","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2021.3066973"},{"key":"e_1_3_2_15_2","first-page":"100087","article-title":"ViT-based terrain recognition system for wearable soft exosuit","volume":"3","author":"Yang F.","year":"2023","unstructured":"F. Yang, C. Chen, Z. Wang, H. Chen, Y. Liu, G. Li, X. Wu, ViT-based terrain recognition system for wearable soft exosuit. Biomim. Intell. Robot. 3, 100087 (2023).","journal-title":"Biomim. Intell. Robot."},{"key":"e_1_3_2_16_2","doi-asserted-by":"crossref","unstructured":"B. Laschowski W. McNally A. Wong J. McPhee \u201cPreliminary design of an environment recognition system for controlling robotic lower-limb prostheses and exoskeletons \u201d in 2019 IEEE 16th International Conference on Rehabilitation Robotics (ICORR) (IEEE 2019) pp. 868\u2013873.","DOI":"10.1109\/ICORR.2019.8779540"},{"key":"e_1_3_2_17_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-022-05191-1"},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","DOI":"10.3390\/s22207913"},{"key":"e_1_3_2_19_2","doi-asserted-by":"publisher","DOI":"10.1126\/scirobotics.adg3705"},{"key":"e_1_3_2_20_2","doi-asserted-by":"publisher","DOI":"10.1109\/TNSRE.2022.3156786"},{"key":"e_1_3_2_21_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbiomech.2021.110820"},{"key":"e_1_3_2_22_2","doi-asserted-by":"crossref","unstructured":"J. Sloboda P. Stegall R. J. McKindles L. Stirling H. C. Siu \u201cUtility of inter-subject transfer learning for wearable-sensor-based joint torque prediction models \u201d in 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) (IEEE 2021) pp. 4901\u20134907.","DOI":"10.1109\/EMBC46164.2021.9630652"},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.3390\/machines9120367"},{"key":"e_1_3_2_24_2","doi-asserted-by":"publisher","DOI":"10.1186\/s12984-023-01232-6"},{"key":"e_1_3_2_25_2","doi-asserted-by":"publisher","DOI":"10.1145\/3380985"},{"key":"e_1_3_2_26_2","doi-asserted-by":"crossref","unstructured":"F. Mu X. Gu Y. Guo B. Lo \u201cUnsupervised domain adaptation for position-independent IMU based gait analysis \u201d in 2020 IEEE SENSORS (IEEE 2020) pp. 1\u20134.","DOI":"10.1109\/SENSORS47125.2020.9278863"},{"key":"e_1_3_2_27_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.smhl.2021.100226"},{"key":"e_1_3_2_28_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.smhl.2023.100431"},{"key":"e_1_3_2_29_2","doi-asserted-by":"crossref","unstructured":"A. Akbari R. Jafari \u201cTransferring activity recognition models for new wearable sensors with deep generative domain adaptation \u201d in Proceedings of the 18th International Conference on Information Processing in Sensor Networks (Association for Computing Machinery 2019) pp. 85\u201396.","DOI":"10.1145\/3302506.3310391"},{"key":"e_1_3_2_30_2","first-page":"8009","article-title":"MotionTransformer: Transferring neural inertial tracking between domains","volume":"33","author":"Chen C.","year":"2019","unstructured":"C. Chen, Y. Miao, C. X. Lu, L. Xie, P. Blunsom, A. Markham, N. Trigoni, MotionTransformer: Transferring neural inertial tracking between domains. Proc. AAAI Conf. Artif. Intell. 33, 8009\u20138016 (2019).","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2007.901024"},{"key":"e_1_3_2_32_2","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1006223"},{"key":"e_1_3_2_33_2","doi-asserted-by":"publisher","DOI":"10.1186\/s12984-020-00663-9"},{"key":"e_1_3_2_34_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.dib.2017.12.022"},{"key":"e_1_3_2_35_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-023-02767-y"},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2023.3311677"},{"key":"e_1_3_2_37_2","doi-asserted-by":"crossref","unstructured":"J.-Y. Zhu T. Park P. Isola A. A. Efros \u201cUnpaired image-to-image translation using cycle-consistent adversarial networks \u201d in 2017 IEEE International Conference on Computer Vision (ICCV) (IEEE 2017) pp. 2242\u20132251.","DOI":"10.1109\/ICCV.2017.244"},{"key":"e_1_3_2_38_2","doi-asserted-by":"publisher","DOI":"10.1109\/JSEN.2021.3095176"},{"key":"e_1_3_2_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/TMRB.2022.3144025"},{"key":"e_1_3_2_40_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-023-02840-6"},{"key":"e_1_3_2_41_2","doi-asserted-by":"publisher","DOI":"10.1126\/science.aal5054"},{"key":"e_1_3_2_42_2","doi-asserted-by":"publisher","DOI":"10.1152\/japplphysiol.00445.2014"},{"key":"e_1_3_2_43_2","doi-asserted-by":"publisher","DOI":"10.1111\/j.2517-6161.1995.tb02031.x"},{"key":"e_1_3_2_44_2","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2024.3388874"},{"key":"e_1_3_2_45_2","doi-asserted-by":"crossref","unstructured":"X. Mao Q. Li H. Xie R. Y. K. Lau Z. Wang S. Paul Smolley \u201cLeast squares generative adversarial networks \u201d in Proceedings of the IEEE International Conference on Computer Vision (ICCV) (IEEE 2017) pp. 2794\u20132802.","DOI":"10.1109\/ICCV.2017.304"},{"key":"e_1_3_2_46_2","unstructured":"Y. Taigman A. Polyak L. Wolf Unsupervised cross-domain image generation. arXiv:1611.02200 [cs.CV] (2016)."},{"key":"e_1_3_2_47_2","doi-asserted-by":"publisher","DOI":"10.1145\/3550299"},{"key":"e_1_3_2_48_2","doi-asserted-by":"publisher","DOI":"10.3390\/s16010115"},{"key":"e_1_3_2_49_2","doi-asserted-by":"crossref","unstructured":"O. Ronneberger P. Fischer T. Brox \u201cU-Net: Convolutional networks for biomedical image segmentation \u201d in Medical Image Computing and Computer-Assisted Intervention\u2013MICCAI 2015 N. Navab J. Hornegger W. M. Wells A. F. Frangi Eds. (Lecture Notes in Computer Science Springer International Publishing 2015) pp. 234\u2013241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_3_2_50_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2019.2920969"},{"key":"e_1_3_2_51_2","doi-asserted-by":"publisher","DOI":"10.1007\/s10618-019-00619-1"},{"key":"e_1_3_2_52_2","unstructured":"S. Bai J. Z. Kolter V. Koltun An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv:1803.01271 [cs.lG] (2018)."},{"key":"e_1_3_2_53_2","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2023.3262164"},{"key":"e_1_3_2_54_2","doi-asserted-by":"crossref","unstructured":"J. Beil I. Ehrenberger C. Scherer C. Mandery T. Asfour \u201cHuman motion classification based on multi-modal sensor data for lower limb exoskeletons \u201d in 2018 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE 2018) pp. 5431\u20135436.","DOI":"10.1109\/IROS.2018.8594110"}],"container-title":["Science Robotics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.science.org\/doi\/pdf\/10.1126\/scirobotics.ads8652","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,19]],"date-time":"2025-11-19T18:58:32Z","timestamp":1763578712000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.science.org\/doi\/10.1126\/scirobotics.ads8652"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,19]]},"references-count":53,"journal-issue":{"issue":"108","published-print":{"date-parts":[[2025,11,19]]}},"alternative-id":["10.1126\/scirobotics.ads8652"],"URL":"https:\/\/doi.org\/10.1126\/scirobotics.ads8652","relation":{},"ISSN":["2470-9476"],"issn-type":[{"value":"2470-9476","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,19]]},"assertion":[{"value":"2024-09-07","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-10-22","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-11-19","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"eads8652"}}