{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T17:52:04Z","timestamp":1785433924396,"version":"3.56.0"},"reference-count":32,"publisher":"MDPI AG","issue":"12","license":[{"start":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T00:00:00Z","timestamp":1781049600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"State Key Laboratory of Robotics and Systems","award":["SKLRS 2025-ZM-02"],"award-info":[{"award-number":["SKLRS 2025-ZM-02"]}]},{"name":"State Key Laboratory of Robotics and Systems","award":["SKLRS202509B"],"award-info":[{"award-number":["SKLRS202509B"]}]},{"award":["SKLRS 2025-ZM-02"],"award-info":[{"award-number":["SKLRS 2025-ZM-02"]}],"id":[{"id":"https:\/\/ror.org\/015m77g16","id-type":"ROR","asserted-by":"publisher"}]},{"award":["SKLRS202509B"],"award-info":[{"award-number":["SKLRS202509B"]}],"id":[{"id":"https:\/\/ror.org\/015m77g16","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Continuous monitoring of lower-limb kinematics in natural environments is essential for gait analysis and rehabilitation but remains challenging due to the limitations of optical systems and the inaccuracy of sparse inertial sensor methods. To address this, we propose a high-precision, minimalist wearable system utilizing only three inertial measurement units placed on the pelvis and shanks. In the data preprocessing stage, engineering modifications are made based on the traditional gradient descent algorithm to implement adaptive channel adjustment on the acceleration and magnetic data of a single IMU, aiming to alleviate the impact of motion acceleration and external magnetic interference on the temporal feature manifold. Subsequently, a pure Transformer neural network is utilized to capture long-range temporal dependencies, reconstructing full lower-limb kinematics without relying on rigid biomechanical assumptions. The model was optimized and deployed on an STM32N647 microcontroller to achieve real-time edge inference with a low latency of approximately 17 ms. Experimental results demonstrate that the proposed method achieves a mean absolute error of 2.41\u00b0 for level walking, significantly outperforming traditional constrained Kalman filter approaches. Furthermore, it maintains high tracking robustness during complex nonlinear movements such as squatting and lunging. In conclusion, this edge-computing-enabled framework provides an accurate, comfortable, and real-time solution for unconstrained human motion capture in daily scenarios.<\/jats:p>","DOI":"10.3390\/s26123706","type":"journal-article","created":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T02:59:31Z","timestamp":1781146771000},"page":"3706","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Real-Time Kinematic Reconstruction of Human Lower Limbs Using a 3-IMU Wearable Sensor Network, Transformer Model, and Deployable Edge Computing"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-2868-2604","authenticated-orcid":false,"given":"Yang","family":"Yu","sequence":"first","affiliation":[{"name":"State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1211-6444","authenticated-orcid":false,"given":"Wei","family":"Dong","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8562-6787","authenticated-orcid":false,"given":"Hui","family":"Dong","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-6062-6803","authenticated-orcid":false,"given":"Wenda","family":"Wang","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1267-4550","authenticated-orcid":false,"given":"Yongzhuo","family":"Gao","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dongmei","family":"Wu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-9529-4107","authenticated-orcid":false,"given":"Weiqi","family":"Lin","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,6,10]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1016\/j.gaitpost.2014.03.189","article-title":"Quantified self and human movement: A review on the clinical impact of wearable sensing and feedback for gait analysis and intervention","volume":"40","author":"Shull","year":"2014","journal-title":"Gait Posture"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"S13","DOI":"10.1016\/j.kjms.2011.08.004","article-title":"Biomechanics of human movement and its clinical applications","volume":"28","author":"Lu","year":"2012","journal-title":"Kaohsiung J. Med. Sci."},{"key":"ref_3","unstructured":"NOKOV (2026, April 09). NOKOV Motion Capture System. Available online: https:\/\/www.nokov.com\/."},{"key":"ref_4","unstructured":"(2026, April 09). Vicon Motion Systems. Available online: https:\/\/www.vicon.com\/."},{"key":"ref_5","unstructured":"Roetenberg, D., Luinge, H., and Slycke, P. (2009). Xsens MVN: Full 6DOF Human Motion Tracking Using Miniature Inertial Sensors, Xsens Motion Technologies. Available online: https:\/\/www.researchgate.net\/publication\/239920367."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Prisco, G., Pirozzi, M.A., Santone, A., Esposito, F., Cesarelli, M., Amato, F., and Donisi, L. (2025). Validity of wearable inertial sensors for gait analysis: A systematic review. Diagnostics, 15.","DOI":"10.3390\/diagnostics15010036"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"6478","DOI":"10.1109\/TMECH.2025.3540069","article-title":"A wearable scissored-pair control moment gyroscopes utilized for reactionless support in human locomotion","volume":"30","author":"Lin","year":"2025","journal-title":"IEEE\/ASME Trans. Mechatron."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1109\/TBME.2012.2223465","article-title":"A novel approach to reducing number of sensing units for wearable gait analysis systems","volume":"60","author":"Salarian","year":"2013","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Hu, X., Yao, C., and Soh, G.S. (2015, January 11\u201314). Performance evaluation of lower limb ambulatory measurement using reduced inertial measurement units and 3R gait model. Proceedings of the IEEE International Conference on Rehabilitation Robotics (ICORR), Singapore.","DOI":"10.1109\/ICORR.2015.7281257"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Joukov, V., Bonnet, V., Karg, M., Venture, G., and Kuli\u0107, D. (2015, January 3\u20135). Rhythmic EKF for pose estimation during gait. Proceedings of the IEEE-RAS International Conference on Humanoid Robots, Seoul, Republic of Korea.","DOI":"10.1109\/HUMANOIDS.2015.7363510"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1293","DOI":"10.1109\/TBME.2020.3026464","article-title":"Estimating Lower Limb Kinematics Using a Reduced Wearable Sensor Count","volume":"68","author":"Sy","year":"2021","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1171","DOI":"10.1109\/34.625129","article-title":"A smoothly constrained Kalman filter","volume":"19","year":"1997","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"2657","DOI":"10.1109\/TBME.2010.2060723","article-title":"Zero-velocity detection\u2014An algorithm evaluation","volume":"57","author":"Skog","year":"2010","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"349","DOI":"10.1111\/cgf.13131","article-title":"Sparse Inertial Poser: Automatic 3D Human Pose Estimation from Sparse IMUs","volume":"36","author":"Marcard","year":"2017","journal-title":"Comput. Graph. Forum"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.gaitpost.2009.09.004","article-title":"Quantification of soft tissue artifact in lower limb human motion analysis: A systematic review","volume":"31","author":"Peters","year":"2010","journal-title":"Gait Posture"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Wang, W., Jiang, W., Yu, Y., Dong, W., Dong, H., Gao, Y., Wu, D., and Lin, W. (2026). An Intelligent Pressurized Thigh Band for Muscular Assistance and Multi-Mode Activity Recognition. Sensors, 26.","DOI":"10.3390\/s26051502"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Lin, W., Dong, H., Gao, Y., Wang, W., Long, Y., He, L., Mao, X., Wu, D., and Dong, W. (2025). A systematic review of locomotion assistance exoskeletons: Prototype development and technical challenges. Technologies, 13.","DOI":"10.3390\/technologies13020069"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/1966394.1966397","article-title":"Motion reconstruction using sparse accelerometer data","volume":"30","author":"Tautges","year":"2011","journal-title":"ACM Trans. Graph."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Wouda, F.J., Giuberti, M., Bellusci, G., and Veltink, P.H. (2016). Estimation of full-body poses using only five inertial sensors: An eager or lazy learning approach?. Sensors, 16.","DOI":"10.3390\/s16122138"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"120","DOI":"10.1016\/j.gaitpost.2007.11.001","article-title":"Predicting lower limb joint kinematics using wearable motion sensors","volume":"28","author":"Findlow","year":"2008","journal-title":"Gait Posture"},{"key":"ref_21","first-page":"1","article-title":"Deep inertial poser: Learning to reconstruct human pose from sparse inertial measurements in real time","volume":"37","author":"Huang","year":"2018","journal-title":"ACM Trans. Graph."},{"key":"ref_22","first-page":"1","article-title":"TransPose: Real-time 3D Human Translation and Pose Estimation with Six Inertial Sensors","volume":"40","author":"Yi","year":"2021","journal-title":"ACM Trans. Graph."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"185","DOI":"10.1016\/j.gaitpost.2020.10.026","article-title":"Lower body kinematics estimation from wearable sensors for walking and running: A deep learning approach","volume":"83","author":"Hernandez","year":"2021","journal-title":"Gait Posture"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Ord\u00f3\u00f1ez, F.J., and Roggen, D. (2016). Deep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition. Sensors, 16.","DOI":"10.3390\/s16010115"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"176507","DOI":"10.1016\/j.cej.2026.176507","article-title":"Bio-inspired magnetically triggered snap-through mechanism for enhanced mechanical-to-electrical energy conversion","volume":"538","author":"Cong","year":"2026","journal-title":"Chem. Eng. J."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Zimmermann, T., Taetz, B., and Bleser, G. (2018). IMU-to-segment assignment and orientation alignment for the lower body using deep learning. Sensors, 18.","DOI":"10.3390\/s18010302"},{"key":"ref_27","unstructured":"Mahmood, N., Ghorbani, N., Troje, N.F., Pons-Moll, G., and Black, M.J. (November, January 27). AMASS: Archive of motion capture as surface shapes. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Seoul, Republic of Korea."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"851","DOI":"10.1145\/2816795.2818013","article-title":"SMPL: A skinned multi-person linear model","volume":"34","author":"Loper","year":"2015","journal-title":"ACM Trans. Graph."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"239","DOI":"10.1016\/j.gaitpost.2016.11.008","article-title":"25 years of lower limb joint kinematics by using inertial and magnetic sensors: A review of methodological approaches","volume":"51","author":"Picerno","year":"2017","journal-title":"Gait Posture"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"607","DOI":"10.1007\/s42235-025-00830-5","article-title":"A fast integrated gait, footstep, and motion planning framework for wheeled-legged robots","volume":"23","author":"Li","year":"2026","journal-title":"J. Bionic Eng."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Li, T., Wang, L., Li, Q., and Liu, T. (2020, January 6\u20139). Lower-Body Walking Motion Estimation Using Only Two Shank-Mounted Inertial Measurement Units (IMUs). Proceedings of the 2020 IEEE\/ASME International Conference on Advanced Intelligent Mechatronics (AIM), Boston, MA, USA.","DOI":"10.1109\/AIM43001.2020.9158961"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Jiang, Y., Ye, Y., Gopinath, D., Won, J., Winkler, A.W., and Liu, C.K. (2022, January 6\u20139). Transformer Inertial Poser: Real-time Human Motion Reconstruction from Sparse IMUs with Simultaneous Terrain Generation. Proceedings of the SA \u201822: SIGGRAPH Asia 2022 Conference Papers, Daegu, Republic of Korea.","DOI":"10.1145\/3550469.3555428"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/12\/3706\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T03:17:03Z","timestamp":1781147823000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/12\/3706"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,10]]},"references-count":32,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["s26123706"],"URL":"https:\/\/doi.org\/10.3390\/s26123706","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,10]]}}}