{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,7]],"date-time":"2026-03-07T19:02:35Z","timestamp":1772910155001,"version":"3.50.1"},"reference-count":47,"publisher":"MDPI AG","issue":"24","license":[{"start":{"date-parts":[[2020,12,12]],"date-time":"2020-12-12T00:00:00Z","timestamp":1607731200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100009389","name":"Promobilia Foundation","doi-asserted-by":"publisher","award":["18200"],"award-info":[{"award-number":["18200"]}],"id":[{"id":"10.13039\/100009389","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004359","name":"Swedish Research Council","doi-asserted-by":"publisher","award":["2018-00750"],"award-info":[{"award-number":["2018-00750"]}],"id":[{"id":"10.13039\/501100004359","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Lower body segment trajectory and gait phase prediction is crucial for the control of assistance-as-needed robotic devices, such as exoskeletons. In order for a powered exoskeleton with phase-based control to determine and provide proper assistance to the wearer during gait, we propose an approach to predict segment trajectories up to 200 ms ahead (angular velocity of the thigh, shank and foot segments) and five gait phases (loading response, mid-stance, terminal stance, preswing and swing), based on collected data from inertial measurement units placed on the thighs, shanks, and feet. The approach we propose is a long-short term memory (LSTM)-based network, a modified version of recurrent neural networks, which can learn order dependence in sequence prediction problems. The algorithm proposed has a weighted discount loss function that places more weight in predicting the next three to five time frames but also contributes to an overall prediction performance for up to 10 time frames. The LSTM model was designed to learn lower limb segment trajectories using training samples and was tested for generalization across participants. All predicted trajectories were strongly correlated with the measured trajectories, with correlation coefficients greater than 0.98. The proposed LSTM approach can also accurately predict the five gait phases, particularly swing phase with 95% accuracy in inter-subject implementation. The ability of the LSTM network to predict future gait trajectories and gait phases can be applied in designing exoskeleton controllers that can better compensate for system delays to smooth the transition between gait phases.<\/jats:p>","DOI":"10.3390\/s20247127","type":"journal-article","created":{"date-parts":[[2020,12,13]],"date-time":"2020-12-13T23:39:36Z","timestamp":1607902776000},"page":"7127","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":81,"title":["Gait Trajectory and Gait Phase Prediction Based on an LSTM Network"],"prefix":"10.3390","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5592-5372","authenticated-orcid":false,"given":"Binbin","family":"Su","sequence":"first","affiliation":[{"name":"KTH MoveAbility Lab, Department of Engineering Mechanics, Royal Institute of Technology, 10044 Stockholm, Sweden"},{"name":"KTH BioMEx Center, Royal Institute of Technology, 10044 Stockholm, Sweden"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5417-5939","authenticated-orcid":false,"given":"Elena M.","family":"Gutierrez-Farewik","sequence":"additional","affiliation":[{"name":"KTH MoveAbility Lab, Department of Engineering Mechanics, Royal Institute of Technology, 10044 Stockholm, Sweden"},{"name":"KTH BioMEx Center, Royal Institute of Technology, 10044 Stockholm, Sweden"},{"name":"Department of Women\u2019s and Children\u2019s Health, Karolinska Institutet, 17177 Stockholm, Sweden"}]}],"member":"1968","published-online":{"date-parts":[[2020,12,12]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1009","DOI":"10.1115\/1.2049333","article-title":"Design of a robotic gait trainer using spring over muscle actuators for ankle stroke rehabilitation","volume":"127","author":"Bharadwaj","year":"2005","journal-title":"J. Biomech. Eng."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"41","DOI":"10.3389\/fnbot.2019.00041","article-title":"Switching assistance for exoskeletons during cyclic motions","volume":"13","author":"Tagliamonte","year":"2019","journal-title":"Front. Neurorobot."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1109\/TNSRE.2019.2950309","article-title":"Gait trajectory and event prediction from state estimation for exoskeletons during gait","volume":"28","author":"Tanghe","year":"2020","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"752","DOI":"10.1007\/s40846-018-0449-z","article-title":"Development of multi-axis motor control systems for lower limb robotic exoskeleton","volume":"39","author":"Pan","year":"2019","journal-title":"J. Med. Biol. Eng."},{"key":"ref_5","first-page":"136","article-title":"The heat of shortening and the dynamic constants of muscle","volume":"126","author":"Hill","year":"1938","journal-title":"Proc. R. Soc. Lond. Ser. B Biol. Sci."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"381","DOI":"10.1115\/1.1392310","article-title":"Dynamic optimization of human walking","volume":"123","author":"Anderson","year":"2001","journal-title":"J. Biomech. Eng."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"1940","DOI":"10.1109\/TBME.2007.901024","article-title":"OpenSim: Open-source software to create and analyze dynamic simulations of movement","volume":"54","author":"Delp","year":"2007","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"907","DOI":"10.1038\/nn1309","article-title":"Optimality principles in sensorimotor control","volume":"7","author":"Todorov","year":"2004","journal-title":"Nat. Neurosci."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"186","DOI":"10.1016\/j.jbiomech.2013.09.032","article-title":"Statistical method for prediction of gait kinematics with Gaussian process regression","volume":"47","author":"Yun","year":"2014","journal-title":"J. Biomech."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"395","DOI":"10.1017\/S0263574717000467","article-title":"Joint trajectory generator for powered orthosis based on gait modelling using PCA and FFT","volume":"36","author":"Melo","year":"2018","journal-title":"Robotica"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1459","DOI":"10.1109\/TASE.2018.2841358","article-title":"Individualized gait pattern generation for sharing lower limb exoskeleton robot","volume":"15","author":"Wu","year":"2018","journal-title":"IEEE Trans. Autom. Sci. Eng."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Glackin, C., Salge, C., Greaves, M., Polani, D., Slavni\u0107, S., Risti\u0107-Durrant, D., Leu, A., and Matja\u010di\u0107, Z. (2014, January 18\u201320). Gait trajectory prediction using Gaussian process ensembles. Proceedings of the 2014 IEEE-RAS International Conference on Humanoid Robots, Madrid, Spain.","DOI":"10.1109\/HUMANOIDS.2014.7041428"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"1236","DOI":"10.1109\/TNSRE.2019.2914095","article-title":"Gaussian process trajectory learning and synthesis of individualized gait motions","volume":"27","author":"Hong","year":"2019","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Klarenbeek, G., Harmanny, R., and Cifola, L. (2017, January 11\u201313). Multi-target human gait classification using LSTM recurrent neural networks applied to micro-Doppler. Proceedings of the 2017 European Radar Conference (EURAD), Nuremberg, Germany.","DOI":"10.23919\/EURAD.2017.8249173"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"128","DOI":"10.1016\/j.gaitpost.2019.09.007","article-title":"Time series classification using a modified LSTM approach from accelerometer-based data: A comparative study for gait cycle detection","volume":"74","author":"Tan","year":"2019","journal-title":"Gait Posture"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Liang, F.Y., Zhong, C.H., Zhao, X., Castro, D.L., Chen, B., Gao, F., and Liao, W.H. (2018, January 12\u201315). Online adaptive and lstm-based trajectory generation of lower limb exoskeletons for stroke rehabilitation. Proceedings of the 2018 IEEE International Conference on Robotics and Biomimetics (ROBIO), Kuala Lumpur, Malaysia.","DOI":"10.1109\/ROBIO.2018.8664778"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Saleh, K., Hossny, M., and Nahavandi, S. (2017, January 16\u201319). Intent prediction of vulnerable road users from motion trajectories using stacked LSTM network. Proceedings of the 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC), Yokohama, Japan.","DOI":"10.1109\/ITSC.2017.8317941"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"369","DOI":"10.1108\/AA-11-2016-155","article-title":"Deep Spatial-Temporal Model for rehabilitation gait: Optimal trajectory generation for knee joint of lower-limb exoskeleton","volume":"37","author":"Liu","year":"2017","journal-title":"Assem. Autom."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Zaroug, A., Lai, D.T., Mudie, K., and Begg, R. (2020). Lower limb kinematics trajectory prediction using long short-term memory neural networks. Front. Bioeng. Biotechnol., 8.","DOI":"10.3389\/fbioe.2020.00362"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Moreira, L., Cerqueira, S.M., Figueiredo, J., Vilas-Boas, J., and Santos, C.P. (2020, January 15\u201317). AI-based reference ankle joint torque trajectory generation for robotic gait assistance: First steps. Proceedings of the 2020 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC), Ponta Delgada, Portugal.","DOI":"10.1109\/ICARSC49921.2020.9096205"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"3370","DOI":"10.1109\/TNNLS.2019.2891257","article-title":"Multistep prediction of dynamic systems with recurrent neural networks","volume":"30","author":"Mohajerin","year":"2019","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"7157","DOI":"10.1038\/s41598-019-43628-2","article-title":"Gait training using a robotic hip exoskeleton improves metabolic gait efficiency in the elderly","volume":"9","author":"Martini","year":"2019","journal-title":"Sci. Rep."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"561","DOI":"10.1177\/0278364906065505","article-title":"Hybrid control of the Berkeley lower extremity exoskeleton (BLEEX)","volume":"25","author":"Kazerooni","year":"2006","journal-title":"Int. J. Robot. Res."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"57","DOI":"10.3389\/fnbot.2019.00057","article-title":"Stance and swing detection based on the angular velocity of lower limb segments during walking","volume":"13","author":"Grimmer","year":"2019","journal-title":"Front. Neurorobot."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"413","DOI":"10.1109\/TNSRE.2014.2337914","article-title":"A novel adaptive, real-time algorithm to detect gait events from wearable sensors","volume":"23","author":"Bejarano","year":"2014","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"88","DOI":"10.1109\/TNSRE.2015.2409123","article-title":"Assessment of foot trajectory for human gait phase detection using wireless ultrasonic sensor network","volume":"24","author":"Qi","year":"2015","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"6891","DOI":"10.3390\/s140406891","article-title":"IMU-based joint angle measurement for gait analysis","volume":"14","author":"Seel","year":"2014","journal-title":"Sensors"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Wu, G., Wang, C., Wu, X., Wang, Z., Ma, Y., and Zhang, T. (2016, January 1\u20133). Gait phase prediction for lower limb exoskeleton robots. Proceedings of the 2016 IEEE International Conference on Information and Automation (ICIA), Ningbo, China.","DOI":"10.1109\/ICInfA.2016.7831791"},{"key":"ref_30","unstructured":"Wu, X., Ma, Y., Yong, X., Wang, C., He, Y., and Li, N. (2019). Locomotion mode identification and gait phase estimation for exoskeletons during continuous multi-locomotion tasks. IEEE Trans. Cogn. Dev. Syst."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1016\/j.neunet.2018.02.017","article-title":"Adaptive Bayesian inference system for recognition of walking activities and prediction of gait events using wearable sensors","volume":"102","year":"2018","journal-title":"Neural Netw."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1093\/ageing\/26.1.15","article-title":"Comfortable and maximum walking speed of adults aged 20\u201379 years: Reference values and determinants","volume":"26","author":"Bohannon","year":"1997","journal-title":"Age Ageing"},{"key":"ref_33","unstructured":"Gage, J.R., Schwartz, M.H., Koop, S.E., and Novacheck, T.F. (2009). The Identification and Treatment of Gait Problems in Cerebral Palsy, John Wiley & Sons."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Su, B., Smith, C., and Gutierrez Farewik, E. (2020). Gait phase recognition using deep convolutional neural network with inertial measurement units. Biosensors, 10.","DOI":"10.3390\/bios10090109"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"6474","DOI":"10.3390\/s140406474","article-title":"Window size impact in human activity recognition","volume":"14","author":"Banos","year":"2014","journal-title":"Sensors"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"3","DOI":"10.3389\/neuro.07.003.2009","article-title":"Extracting kinematic parameters for monkey bipedal walking from cortical neuronal ensemble activity","volume":"3","author":"Fitzsimmons","year":"2009","journal-title":"Front. Integr. Neurosci."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"212","DOI":"10.1109\/TNSRE.2012.2188304","article-title":"Decoding intra-limb and inter-limb kinematics during treadmill walking from scalp electroencephalographic (EEG) signals","volume":"20","author":"Presacco","year":"2012","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_38","unstructured":"Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep Learning, MIT Press. Available online: http:\/\/www.deeplearningbook.org."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"335","DOI":"10.2106\/00004623-196446020-00009","article-title":"Walking patterns of normal men","volume":"46","author":"Murray","year":"1964","journal-title":"J. Bone Jt. Surg."},{"key":"ref_40","unstructured":"Wang, L., van Asseldonk, E.H., and van der Kooij, H. (July, January 29). Model predictive control-based gait pattern generation for wearable exoskeletons. Proceedings of the 2011 IEEE International Conference on Rehabilitation Robotics, Zurich, Switzerland."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"379","DOI":"10.1109\/TNSRE.2007.903919","article-title":"Design and evaluation of the LOPES exoskeleton robot for interactive gait rehabilitation","volume":"15","author":"Veneman","year":"2007","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Bae, J., Siviy, C., Rouleau, M., Menard, N., O\u2019Donnell, K., Geliana, I., Athanassiu, M., Ryan, D., Bibeau, C., and Sloot, L. (2018, January 21\u201325). A lightweight and efficient portable soft exosuit for paretic ankle assistance in walking after stroke. Proceedings of the 2018 IEEE International Conference on Robotics and Automation (ICRA), Brisbane, QLD, Australia.","DOI":"10.1109\/ICRA.2018.8461046"},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"39","DOI":"10.3389\/fnbot.2018.00039","article-title":"Leg force control through biarticular muscles for human walking assistance","volume":"12","author":"Sharbafi","year":"2018","journal-title":"Front. Neurorobot."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"1650005","DOI":"10.1142\/S0219843616500055","article-title":"A subject-specific EMG-driven musculoskeletal model for applications in lower-limb rehabilitation robotics","volume":"13","author":"Ai","year":"2016","journal-title":"Int. J. Humanoid Robot."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"18","DOI":"10.3389\/fnbot.2018.00018","article-title":"A subject-specific kinematic model to predict human motion in exoskeleton-assisted gait","volume":"12","author":"Torricelli","year":"2018","journal-title":"Front. Neurorobot."},{"key":"ref_46","unstructured":"Jozefowicz, R., Zaremba, W., and Sutskever, I. (2015, January 6\u201311). An empirical exploration of recurrent network architectures. Proceedings of the International Conference on Machine Learning, Lille, France."},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Desai, J.P. (2018). Encyclopedia Of Medical Robotics, The (In 4 Volumes), World Scientific.","DOI":"10.1142\/10770-vol4"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/20\/24\/7127\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T10:44:12Z","timestamp":1760179452000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/20\/24\/7127"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,12,12]]},"references-count":47,"journal-issue":{"issue":"24","published-online":{"date-parts":[[2020,12]]}},"alternative-id":["s20247127"],"URL":"https:\/\/doi.org\/10.3390\/s20247127","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,12,12]]}}}