{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T12:26:35Z","timestamp":1784204795039,"version":"3.55.0"},"reference-count":36,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T00:00:00Z","timestamp":1779235200000},"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:p>Active physical human\u2013exoskeleton interaction has been widely studied. However, the challenges of human motion intention recognition and synchronous tracking have not been well-addressed. In this article, a motion intention recognition method based on biophysical information fusion and adaptive learning was proposed to overcome the limitations of existing approaches. First, a lower-limb joint angle prediction model was developed by integrating surface electromyography (sEMG), historical joint angles and centers of gravity. The convolutional neural network, Mamba network, and multilayer perceptron network were used respectively for feature extraction, information fusion, and joint angle prediction. Second, an online adaptive method for the angle prediction model was designed based on a style transfer mapping technique to address the issue of recognition accuracy decline. In this method, the new sEMG features were mapped into the initial feature space, by which the prediction model can maintain the predictive performance during long-term implementation. Furthermore, a real-time control method for the exoskeleton synchronous tracking was given based on the predicted angles. Finally, the feasibility of the proposed methods was validated through the offline and online experiments.<\/jats:p>","DOI":"10.3389\/fnbot.2026.1779125","type":"journal-article","created":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T05:48:29Z","timestamp":1779256109000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Active physical human-exoskeleton interaction based on motion intention adaptive recognition and synchronous tracking"],"prefix":"10.3389","volume":"20","author":[{"given":"Weiguo","family":"Shi","sequence":"first","affiliation":[{"name":"School of Artificial Intelligence, University of Chinese Academy of Sciences","place":["Beijing, China"]},{"name":"State Key Laboratory of Multimodal Artificial Intelligence Systems","place":["Institute of Automation, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weiqun","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, University of Chinese Academy of Sciences","place":["Beijing, China"]},{"name":"State Key Laboratory of Multimodal Artificial Intelligence Systems","place":["Institute of Automation, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiaxing","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, University of Chinese Academy of Sciences","place":["Beijing, China"]},{"name":"State Key Laboratory of Multimodal Artificial Intelligence Systems","place":["Institute of Automation, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shengda","family":"Liu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Multimodal Artificial Intelligence Systems","place":["Institute of Automation, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xu","family":"Liang","sequence":"additional","affiliation":[{"name":"School of Automation and Intelligence, Beijing Jiaotong University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tianyu","family":"Lin","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, University of Chinese Academy of Sciences","place":["Beijing, China"]},{"name":"State Key Laboratory of Multimodal Artificial Intelligence Systems","place":["Institute of Automation, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kexin","family":"Xiang","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, University of Chinese Academy of Sciences","place":["Beijing, China"]},{"name":"State Key Laboratory of Multimodal Artificial Intelligence Systems","place":["Institute of Automation, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zeng-Guang","family":"Hou","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, University of Chinese Academy of Sciences","place":["Beijing, China"]},{"name":"State Key Laboratory of Multimodal Artificial Intelligence Systems","place":["Institute of Automation, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,20]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"370","DOI":"10.1109\/TNSRE.2019.2962189","article-title":"A deep transfer learning approach to reducing the effect of electrode shift in emg pattern recognition-based control","volume":"28","author":"Ameri","year":"2020","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1068","DOI":"10.1109\/TNSRE.2021.3086401","article-title":"Inter-subject domain adaptation for cnn-based wrist kinematics estimation using semg","volume":"29","author":"Bao","year":"2021","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng"},{"key":"B3","doi-asserted-by":"publisher","first-page":"327","DOI":"10.1016\/j.clinbiomech.2009.01.010","article-title":"Surface emg based muscle fatigue evaluation in biomechanics","volume":"24","author":"Cifrek","year":"2009","journal-title":"Clin. Biomech"},{"key":"B4","article-title":"\u201cTransformers are SSMs: generalized models and efficient algorithms through structured state space duality,\u201d","volume-title":"Proceedings of the 41st international conference on machine learning","author":"Dao","year":"2024"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/S0169-7439(99)00047-7","article-title":"The mahalanobis distance","volume":"50","author":"De Maesschalck","year":"2000","journal-title":"Chemometr. Intell. Lab. Syst"},{"key":"B6","doi-asserted-by":"crossref","first-page":"700","DOI":"10.1109\/EMBC46164.2021.9630609","article-title":"\u201cEstimation of joint angle from sEMG and inertial measurements based on deep learning approach,\u201d","volume-title":"2021 43rd annual international conference of the IEEE engineering in medicine & biology society (EMBC)","author":"Delgado","year":"2021"},{"key":"B7","article-title":"\u201cMamba: linear-time sequence modeling with selective state spaces,\u201d","volume-title":"The first conference on language modeling (COLM 2024)","author":"Gu","year":"2024"},{"key":"B8","first-page":"1","article-title":"\u201cCNN-LSTM network based prediction of human joint angles using multi-band sEMG and historical angles,\u201d","volume-title":"Proceedings of the 2021 international joint conference on neural networks (IJCNN)","author":"Jiao","year":"2021"},{"key":"B9","doi-asserted-by":"publisher","first-page":"100","DOI":"10.1038\/s41597-024-04296-8","article-title":"Emg dataset for gesture recognition with arm translation","volume":"12","author":"Kyranou","year":"2025","journal-title":"Sci. Data"},{"key":"B10","doi-asserted-by":"publisher","first-page":"3281","DOI":"10.1109\/TCYB.2019.2904052","article-title":"Multisource transfer learning for cross-subject eeg emotion recognition","volume":"50","author":"Li","year":"","journal-title":"IEEE Trans. Cybernet"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12938-019-0653-2","article-title":"Estimation of continuous elbow joint movement based on human physiological structure","volume":"18","author":"Li","year":"2019","journal-title":"Biomed. Eng. Online"},{"key":"B12","doi-asserted-by":"publisher","first-page":"102074","DOI":"10.1016\/j.bspc.2020.102074","article-title":"A review of the key technologies for semg-based human-robot interaction systems","volume":"62","author":"Li","year":"","journal-title":"Biomed. Sig. Process. Control"},{"key":"B13","doi-asserted-by":"publisher","first-page":"43","DOI":"10.3390\/electronics9010043","article-title":"Estimation of knee movement from surface emg using random forest with principal component analysis","volume":"9","author":"Li","year":"","journal-title":"Electronics"},{"key":"B14","doi-asserted-by":"publisher","first-page":"6999","DOI":"10.1109\/TNNLS.2021.3084827","article-title":"A survey of convolutional neural networks: analysis, applications, and prospects","volume":"33","author":"Li","year":"2022","journal-title":"IEEE Trans. Neural Netw. Learn. Syst"},{"key":"B15","doi-asserted-by":"publisher","first-page":"26","DOI":"10.1109\/THMS.2021.3129708","article-title":"Finite-time observer based variable impedance control of cable-driven continuum manipulators","volume":"52","author":"Liang","year":"2022","journal-title":"IEEE Trans. Hum.-Mach. Syst"},{"key":"B16","doi-asserted-by":"publisher","first-page":"4814","DOI":"10.1109\/TMECH.2024.3394491","article-title":"Adaptive human-robot interaction torque estimation with high accuracy and strong tracking ability for a lower limb rehabilitation robot","volume":"29","author":"Liang","year":"2024","journal-title":"IEEE\/ASME Trans. Mechatr"},{"key":"B17","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1038\/s41586-023-06094-5","article-title":"Walking naturally after spinal cord injury using a brain-spine interface","volume":"618","author":"Lorach","year":"2023","journal-title":"Nature"},{"key":"B18","doi-asserted-by":"publisher","first-page":"5072","DOI":"10.1109\/TASE.2024.3415650","article-title":"A physical human-robot interaction framework for trajectory adaptation based on human motion prediction and adaptive impedance control","volume":"22","author":"Luo","year":"2024","journal-title":"IEEE Trans. Autom. Sci. Eng"},{"key":"B19","doi-asserted-by":"publisher","first-page":"925","DOI":"10.1109\/TNSRE.2025.3541227","article-title":"High-density emg grip force estimation during muscle fatigue via domain adaptation","volume":"33","author":"Pan","year":"2025","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng"},{"key":"B20","doi-asserted-by":"publisher","first-page":"1202","DOI":"10.1109\/TMRB.2024.3408896","article-title":"Enhanced emg-based hand gesture classification in real-world scenarios: mitigating dynamic factors with tempo-spatial wavelet transform and deep learning","volume":"6","author":"Rani","year":"2024","journal-title":"IEEE Trans. Med. Robot. Bion"},{"key":"B21","doi-asserted-by":"publisher","first-page":"528","DOI":"10.1109\/TMRB.2023.3292451","article-title":"Continuous prediction of human joint mechanics using emg signals: a review of model-based and model-free approaches","volume":"5","author":"Sitole","year":"2023","journal-title":"IEEE Trans. Med. Robot. Bion"},{"key":"B22","doi-asserted-by":"publisher","first-page":"107124","DOI":"10.1016\/j.compbiomed.2023.107124","article-title":"Continuous online prediction of lower limb joints angles based on semg signals by deep learning approach","volume":"163","author":"Song","year":"2023","journal-title":"Comput. Biol. Med"},{"key":"B23","doi-asserted-by":"publisher","first-page":"226","DOI":"10.1109\/TNSRE.2015.2413393","article-title":"Detection of and compensation for emg disturbances for powered lower limb prosthesis control","volume":"24","author":"Spanias","year":"2015","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng"},{"key":"B24","doi-asserted-by":"publisher","first-page":"4130","DOI":"10.1109\/TASE.2026.3653476","article-title":"Trajectory Planning Method for Fracture Reduction of Parallel Robots Based on DMP and APF","volume":"23","author":"Su","year":"2026","journal-title":"IEEE Trans. Autom. Sci. Eng."},{"key":"B25","doi-asserted-by":"publisher","first-page":"1605","DOI":"10.1109\/TNSRE.2020.2991643","article-title":"Understanding limb position and external load effects on real-time pattern recognition control in amputees","volume":"28","author":"Teh","year":"2020","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng"},{"key":"B26","doi-asserted-by":"publisher","first-page":"106162","DOI":"10.1016\/j.bspc.2024.106162","article-title":"Data-driven estimation for uphill continuous rehabilitation motion at different slopes using semg","volume":"93","author":"Tian","year":"2024","journal-title":"Biomed. Sig. Process. Control"},{"key":"B27","article-title":"Neural correlates of single-task versus cognitive-motor dual-task training.","author":"Wang","year":"2021"},{"key":"B36","doi-asserted-by":"publisher","first-page":"532","DOI":"10.1109\/TCDS.2021.3053050","volume":"14","year":"","journal-title":"IEEE Trans. Cogn. Dev. Syst"},{"key":"B28","doi-asserted-by":"publisher","first-page":"343","DOI":"10.1109\/TRO.2022.3189231","article-title":"Drivable space of rehabilitation robot for physical human-robot interaction: definition and an expanding method","volume":"39","author":"Wang","year":"2023","journal-title":"IEEE Trans. Robot"},{"key":"B29","doi-asserted-by":"publisher","first-page":"4087","DOI":"10.1109\/TMECH.2024.3362381","article-title":"A multiposture robot for full cycle rehabilitation of lower limbs: design and autonomous training","volume":"29","author":"Wang","year":"2024","journal-title":"IEEE\/ASME Trans. Mechatr"},{"key":"B30","doi-asserted-by":"publisher","first-page":"1487","DOI":"10.1109\/TNSRE.2024.3383857","article-title":"Continuous motion intention prediction using semg for upper-limb rehabilitation: a systematic review of model-based and model-free approaches","volume":"32","author":"Wei","year":"2024","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng"},{"key":"B31","doi-asserted-by":"publisher","first-page":"13019","DOI":"10.1109\/JSEN.2021.3068521","article-title":"Advances and disturbances in semg-based intentions and movements recognition: a review","volume":"21","author":"Xu","year":"2021","journal-title":"IEEE Sens. J"},{"key":"B32","doi-asserted-by":"publisher","first-page":"1718","DOI":"10.1109\/JBHI.2021.3122277","article-title":"Fatigue-sensitivity comparison of semg and a-mode ultrasound based hand gesture recognition","volume":"26","author":"Zeng","year":"2022","journal-title":"IEEE J. Biomed. Health Inform"},{"key":"B33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2022.3227604","article-title":"Boosting personalized musculoskeletal modeling with physics-informed knowledge transfer","volume":"72","author":"Zhang","year":"2022","journal-title":"IEEE Trans. Instrument. Measur"},{"key":"B34","first-page":"393","article-title":"\u201cStyle transfer matrix learning for writer adaptation,\u201d","volume-title":"2011 IEEE conference on computer vision and pattern recognition (CVPR)","author":"Zhang","year":"2011"},{"key":"B35","doi-asserted-by":"publisher","first-page":"1252","DOI":"10.1109\/TNSRE.2022.3173708","article-title":"Senic: an open source dataset for semg-based gesture recognition in non-ideal conditions","volume":"30","author":"Zhu","year":"2022","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2026.1779125\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T05:48:32Z","timestamp":1779256112000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2026.1779125\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,20]]},"references-count":36,"alternative-id":["10.3389\/fnbot.2026.1779125"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2026.1779125","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,20]]},"article-number":"1779125"}}