{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T22:44:55Z","timestamp":1790117095777,"version":"4.0.1"},"reference-count":50,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2025,6,3]],"date-time":"2025-06-03T00:00:00Z","timestamp":1748908800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"University of Canberra (University of Canberra Higher Degree by Research Stipend Scholarship (UC-HDR)","award":["SCH-001512\/MXP001"],"award-info":[{"award-number":["SCH-001512\/MXP001"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Upper limb assistive exoskeletons help stroke patients by assisting arm movement in impaired individuals. However, effective control of these systems to help stroke survivors is a complex task. In this paper, a novel approach is proposed to enhance the control of upper limb assistive exoskeletons by using torque estimation and prediction in a proportional\u2013integral\u2013derivative (PID) controller loop to more optimally integrate the torque of the exoskeleton robot, which aims to eliminate system uncertainties. First, a model for torque estimation from Electromyography (EMG) signals and a predictive torque model for the upper limb exoskeleton robot for the elbow are trained. The trained data consisted of two-dimensional high-density surface EMG (HD-sEMG) signals to record myoelectric activity from five upper limb muscles (biceps brachii, triceps brachii, anconeus, brachioradialis, and pronator teres) during voluntary isometric contractions for twelve healthy subjects performing four different isometric tasks (supination\/pronation and elbow flexion\/extension) for one minute each, which were trained on long short-term memory (LSTM), bidirectional LSTM (BLSTM), and gated recurrent units (GRU) deep neural network models. These models estimate and predict torque requirements. Finally, the estimated and predicted torque from the trained network is used online as input to a PID control loop and robot dynamic, which aims to control the robot optimally. The results showed that using the proposed method creates a strong and innovative approach to greater independence and rehabilitation improvement.<\/jats:p>","DOI":"10.3390\/s25113528","type":"journal-article","created":{"date-parts":[[2025,6,3]],"date-time":"2025-06-03T11:55:52Z","timestamp":1748951752000},"page":"3528","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":11,"title":["Enhancing Upper Limb Exoskeletons Using Sensor-Based Deep Learning Torque Prediction and PID Control"],"prefix":"10.3390","volume":"25","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7211-4619","authenticated-orcid":false,"given":"Farshad","family":"Shakeriaski","sequence":"first","affiliation":[{"name":"Faculty of Science and Technology, University of Canberra, Canberra 2617, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Masoud","family":"Mohammadian","sequence":"additional","affiliation":[{"name":"Faculty of Science and Technology, University of Canberra, Canberra 2617, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,6,3]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1944","DOI":"10.4103\/NRR.NRR-D-24-00392","article-title":"Beyond wrecking a wall: Revisiting the concept of blood\u2013brain barrier breakdown in ischemic stroke","volume":"20","year":"2025","journal-title":"Neural Regen. Res."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"327","DOI":"10.1161\/01.STR.2.4.327","article-title":"Cerebral apoplexy (stroke): Pathogenesis, pathophysiology and therapy as illustrated by regional blood flow measurements in the brain","volume":"2","author":"Paulson","year":"1971","journal-title":"Stroke"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"437","DOI":"10.1007\/s12975-017-0593-8","article-title":"Photothrombotic stroke as a model of ischemic stroke","volume":"9","author":"Uzdensky","year":"2018","journal-title":"Transl. Stroke Res."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"668","DOI":"10.1177\/1747493019847334","article-title":"Canadian stroke best practice recommendations: Mood, cognition and fatigue following stroke, update 2019","volume":"15","author":"Lindsay","year":"2020","journal-title":"Int. J. Stroke"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Pei, D., and Vinjamuri, R. (2025). Investigating the role of cortical rhythms in modulating kinematic synergies and exploring their potential for stroke rehabilitation. Brain-Computer Interfaces, Elsevier.","DOI":"10.1016\/B978-0-323-95439-6.00019-3"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Syringas, P., Potsika, V., Tachos, N., Pardalis, A., Papaioannou, C., Mitsis, A., Pakos, E.E., Zestas, O.N., Papagiannis, G., and Triantafyllou, A. (2025). Exploring New Tools in Upper Limb Rehabilitation After Stroke Using an Exoskeletal Aid: A Pilot Randomized Control Study. Healthcare, 13.","DOI":"10.3390\/healthcare13010091"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Jin, C., Chen, Y., and Ma, Y. (2025). Effectiveness of robot-assisted task-oriented training intervention for upper limb and daily living skills in stroke patients: A meta-analysis. PLoS ONE, 20.","DOI":"10.1371\/journal.pone.0316633"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"2229","DOI":"10.1007\/s10072-023-06723-x","article-title":"Evaluation of an upper limb robotic rehabilitation program on motor functions, quality of life, cognition, and emotional status in patients with stroke: A randomized controlled study","volume":"44","author":"Taravati","year":"2022","journal-title":"Neurol. Sci."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Chiyohara, S., Furukawa, J.-i., Noda, T., Morimoto, J., and Imamizu, H. (2020). Passive training with upper extremity exoskeleton robot affects proprioceptive acuity and performance of motor learning. Sci. Rep., 10.","DOI":"10.1038\/s41598-020-68711-x"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"14","DOI":"10.1186\/s12984-022-00991-y","article-title":"A randomized clinical control study on the efficacy of three-dimensional upper limb robotic exoskeleton training in chronic stroke","volume":"19","author":"Frisoli","year":"2022","journal-title":"J. NeuroEng. Rehabil."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s41133-020-00043-x","article-title":"Development of robot-based upper limb devices for rehabilitation purposes: A systematic review","volume":"6","author":"Narayan","year":"2021","journal-title":"Augment. Hum. Res."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1310\/G8XB-VM23-1TK7-PWQU","article-title":"Robotic technology and stroke rehabilitation: Translating research into practice","volume":"11","author":"Fasoli","year":"2004","journal-title":"Top. Stroke Rehabil."},{"key":"ref_13","unstructured":"Hogan, N., Krebs, H.I., Charnnarong, J., Srikrishna, P., and Sharon, A. (1992, January 1\u20133). MIT-MANUS: A workstation for manual therapy and training. I. Proceedings of the IEEE International Workshop on Robot and Human Communication, Tokyo, Japan."},{"key":"ref_14","first-page":"653","article-title":"Understanding and treating arm movement impairment after chronic brain injury: Progress with the ARM guide","volume":"37","author":"Reinkensmeyer","year":"2014","journal-title":"J. Rehabil. Res. Dev."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"560","DOI":"10.1109\/TNSRE.2007.908560","article-title":"Design, implementation and clinical tests of a wire-based robot for neurorehabilitation","volume":"15","author":"Rosati","year":"2007","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Song, L., Ju, C., Cui, H., Qu, Y., Xu, X., and Chen, C. (2025). Research on Control Strategy Technology of Upper Limb Exoskeleton Robots. Machines, 13.","DOI":"10.3390\/machines13030207"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"197","DOI":"10.1109\/THMS.2025.3529759","article-title":"Fuzzy Adaptive Controller of a Wearable Assistive Upper Limb Exoskeleton Using a Disturbance Observer","volume":"55","author":"Amiri","year":"2025","journal-title":"IEEE Trans. Hum.-Mach. Syst."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Wang, C., Peng, L., and Hou, Z.-G. (2025). Development and control of an upper-limb exoskeleton casia-exo for motor learning in post-stroke rehabilitation. IEEE\/CAA J. Autom. Sin., 1\u20133.","DOI":"10.1109\/JAS.2024.124662"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Benos, L., Tsaopoulos, D., Tagarakis, A.C., Kateris, D., Busato, P., and Bochtis, D. (2025). Explainable AI-Enhanced Human Activity Recognition for Human\u2013Robot Collaboration in Agriculture. Appl. Sci., 15.","DOI":"10.3390\/app15020650"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"673","DOI":"10.1109\/TSMCB.2012.2214381","article-title":"Neural PID control of robot manipulators with application to an upper limb exoskeleton","volume":"43","author":"Yu","year":"2013","journal-title":"IEEE Trans. Cybern."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"1012","DOI":"10.1109\/LRA.2021.3056366","article-title":"Sensor reduction, estimation, and control of an upper-limb exoskeleton","volume":"6","author":"Sun","year":"2021","journal-title":"IEEE Rob. Autom. Lett."},{"key":"ref_22","unstructured":"Lee, B.-K., Lee, H.-D., Lee, J.-y., Shin, K., Han, J.-S., and Han, C.-S. (2012, January 14\u201318). Development of dynamic model-based controller for upper limb exoskeleton robot. Proceedings of the 2012 IEEE International Conference on Robotics and Automation, Saint Paul, MN, USA."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"844","DOI":"10.1016\/j.isatra.2013.05.003","article-title":"Adaptive control of 5 DOF upper-limb exoskeleton robot with improved safety","volume":"52","author":"Kang","year":"2013","journal-title":"ISA Trans."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Alam, N., Hasan, S., Mashud, G.A., and Bhujel, S. (2025). Neural Network for Enhancing Robot-Assisted Rehabilitation: A Systematic Review. Actuators, 14.","DOI":"10.3390\/act14010016"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"e13659","DOI":"10.1111\/exsy.13659","article-title":"Convolution-enhanced vision transformer method for lower limb exoskeleton locomotion mode recognition","volume":"41","author":"Zheng","year":"2024","journal-title":"Expert Syst."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Ren, J., Wang, A., Li, H., Yue, X., and Meng, L. (2023). A transformer-based neural network for gait prediction in lower limb exoskeleton robots using plantar force. Sensors, 23.","DOI":"10.3390\/s23146547"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"82730","DOI":"10.1109\/ACCESS.2024.3413576","article-title":"LLMT: A transformer-based multi-modal lower limb human motion prediction model for assistive robotics applications","volume":"12","author":"Hosseini","year":"2024","journal-title":"IEEE Access"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Zeng, M., Gu, J., and Feng, Y. (2023, January 8\u201310). Motion Prediction Based on sEMG-Transformer for Lower Limb Exoskeleton Robot Control. Proceedings of the 2023 International Conference on Advanced Robotics and Mechatronics (ICARM), Sanya, China.","DOI":"10.1109\/ICARM58088.2023.10218920"},{"key":"ref_29","first-page":"2581","article-title":"An upper-limb exoskeleton robot control using a novel fast fuzzy sliding mode control","volume":"36","author":"Rahmani","year":"2019","journal-title":"J. Intell. Fuzzy Syst."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"397","DOI":"10.1038\/s41597-020-00717-6","article-title":"High-density surface electromyography signals during isometric contractions of elbow muscles of healthy humans","volume":"7","author":"Serna","year":"2020","journal-title":"Sci. Data"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2015, January 7\u201313). Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. Proceedings of the IEEE International Conference on Computer Vision, Las Condes, Chile.","DOI":"10.1109\/ICCV.2015.123"},{"key":"ref_32","unstructured":"Saxe, A.M., McClelland, J.L., and Ganguli, S. (2013). Exact solutions to the nonlinear dynamics of learning in deep linear neural networks. arXiv."},{"key":"ref_33","unstructured":"Patil, S.R., and Seth, D. (July, January 29). Decoding Elbow Movement Intentions from EMG Signals for Exosuit\/Exoskeleton Control. Proceedings of the International Conference on Human-Computer Interaction, Washinton, DC, USA."},{"key":"ref_34","unstructured":"Huang, Z., Xu, W., and Yu, K. (2015). Bidirectional LSTM-CRF models for sequence tagging. arXiv."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s42600-024-00382-3","article-title":"Evaluation of long-short-term memory on simultaneous and proportional myoelectric control system for individual finger movements","volume":"41","author":"Anam","year":"2025","journal-title":"Res. Biomed. Eng."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Dey, R., and Salem, F.M. (2017, January 6\u20139). Gate-variants of gated recurrent unit (GRU) neural networks. Proceedings of the 2017 IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS), Boston, MA, USA.","DOI":"10.1109\/MWSCAS.2017.8053243"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Cheng, G., Povey, D., Huang, L., Xu, J., Khudanpur, S., and Yan, Y. (2018, January 2\u20136). Output-Gate Projected Gated Recurrent Unit for Speech Recognition. Proceedings of the Interspeech 2018, Hyderabad, India.","DOI":"10.21437\/Interspeech.2018-1403"},{"key":"ref_38","unstructured":"Glorot, X., and Bengio, Y. (2010, January 13\u201315). Understanding the difficulty of training deep feedforward neural networks. Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, Sardinia, Italy."},{"key":"ref_39","first-page":"12360","article-title":"Root mean square layer normalization","volume":"32","author":"Zhang","year":"2019","journal-title":"Adv. Neural. Inf. Process. Syst."},{"key":"ref_40","first-page":"2413","article-title":"Pearson correlation coefficient-based performance enhancement of broad learning system for stock price prediction","volume":"69","author":"Li","year":"2022","journal-title":"IEEE Trans. Circuits Syst. II Express Briefs"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Narayan, J., Kalani, A., and Dwivedy, S.K. (2019, January 10\u201312). Reference trajectory based Jacobian transpose control of a novel lower limb exoskeleton system for children. Proceedings of the 2019 5th International Conference on Signal Processing, Computing and Control (ISPCC), Solan, India.","DOI":"10.1109\/ISPCC48220.2019.8988474"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Ahmed, T., Islam, M.R., Brahmi, B., and Rahman, M.H. (2022). Robustness and tracking performance evaluation of PID motion control of 7 DoF anthropomorphic exoskeleton robot assisted upper limb rehabilitation. Sensors, 22.","DOI":"10.3390\/s22103747"},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"106505","DOI":"10.1016\/j.ijepes.2020.106505","article-title":"A novel LMSRE-based adaptive PI control scheme for grid-integrated PMSG-based variable-speed wind turbine","volume":"125","author":"Qais","year":"2021","journal-title":"Int. J. Electr. Power Energy Syst."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Foroutannia, A., Akbarzadeh-T, M.-R., and Akbarzadeh, A. (2022). A deep learning strategy for EMG-based joint position prediction in hip exoskeleton assistive robots. Biomed. Signal Process. Control, 75.","DOI":"10.1016\/j.bspc.2022.103557"},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"102952","DOI":"10.1016\/j.mechatronics.2023.102952","article-title":"Adaptive fuzzy impedance control of exoskeleton robots with electromyography-based convolutional neural networks for human intended trajectory estimation","volume":"91","author":"Foroutannia","year":"2023","journal-title":"Mechatronics"},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"174940","DOI":"10.1109\/ACCESS.2019.2956951","article-title":"A continuous estimation model of upper limb joint angles by using surface electromyography and deep learning method","volume":"7","author":"Chen","year":"2019","journal-title":"IEEE Access"},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Baby Jephil, P., Acharaya, P., Xu, L., Guo, K., Yu, H., Watsford, M., Rong, S., and Su, S. (2020). Estimation of ankle joint torque and angle based on S-EMG signal for assistive rehabilitation robots. Biomedical Signal Processing: Advances in Theory, Algorithms and Applications, Springer.","DOI":"10.1007\/978-981-13-9097-5_2"},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"Hu, B., Mao, B., Lu, S., and Yu, H. (2022). Design and torque control base on neural network PID of a variable stiffness joint for rehabilitation robot. Front. Neurorobot., 16.","DOI":"10.3389\/fnbot.2022.1007324"},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"16021","DOI":"10.1007\/s00521-021-06785-y","article-title":"AI-driven rehabilitation and assistive robotic system with intelligent PID controller based on RBF neural networks","volume":"35","author":"Xiao","year":"2023","journal-title":"Neural Comput. Appl."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"2289","DOI":"10.1109\/TBME.2023.3240879","article-title":"Accurate real-time joint torque estimation for dynamic prediction of human locomotion","volume":"70","author":"Dinovitzer","year":"2023","journal-title":"IEEE Trans. Biomed. Eng."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/25\/11\/3528\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T17:46:54Z","timestamp":1760032014000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/25\/11\/3528"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,3]]},"references-count":50,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2025,6]]}},"alternative-id":["s25113528"],"URL":"https:\/\/doi.org\/10.3390\/s25113528","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,3]]}}}