Upper limb motor dysfunction caused by stroke, traumatic brain injury, and other neurological disorders has become a major global public health challenge, imposing heavy burdens on both healthcare systems and patient families. Conventional manual rehabilitation therapy is limited by insufficient therapist resources, inconsistent treatment quality, and a lack of quantitative evaluation, while most existing rehabilitation robots suffer from low terminal accuracy, poor adaptability to individual patient status, and low long-term training compliance. This paper presents a hybrid parallel-serial upper limb rehabilitation robot integrating multi-modal force-tactile perception, intelligent adaptive trajectory control, and immersive augmented reality (AR) interaction. First, a multi-body dynamic model of the hybrid mechanism is established based on the Lagrangian method, and a domestically manufactured hardware platform is developed with 100% localization rate of core components, achieving 0.1 mm terminal repeat positioning accuracy. Second, a multi-modal sensing fusion framework combining surface electromyography (sEMG), 3D force signals, and kinematic data is constructed; an optimized BiLSTM-Attention model with reinforcement learning is proposed to predict patient motion intention 300 ms in advance, and dynamically adjust auxiliary torque and motion trajectory with a control response delay ≤ 30 ms. Third, a lightweight asynchronous AR training system is developed to realize haptic-visual synchronous feedback, with interaction latency controlled within 50 ms. Bench tests and preliminary clinical validation show that the proposed robot outperforms industry average levels in core performance indicators, increases patients’ active motion induction rate by 30%, and shortens the Brunnstrom stage progression cycle by approximately 40% compared with traditional manual training. This system provides a precise, intelligent, and patient-friendly solution for upper limb motor rehabilitation, with broad clinical application prospects.
References
[1]
Phan, H.T., Reeves, M.J., Gall, S., Morgenstern, L.B., Xu, Y. and Lisabeth, L.D. (2022) Factors Contributing to Sex Differences in Health-Related Quality of Life after Ischemic Stroke: BASIC (Brain Attack Surveillance in Corpus Christi) Project. Journal of the American Heart Association, 11, e026123.
https://doi.org/10.1161/jaha.122.026123
[2]
Morgenstern, L.B., Hollenhorst, C.N., Gallo, L.C., Hsu, C., Kim, S., Gibbs, R., et al. (2021) Ethnic Differences in Informal Caregiving after Stroke. Stroke, 52, 3938-3943.
https://doi.org/10.1161/strokeaha.120.032740
[3]
Xing, Y. and Bai, Y. (2020) A Review of Exercise-Induced Neuroplasticity in Ischemic Stroke: Pathology and Mechanisms. Molecular Neurobiology, 57, 4218-4231.
https://doi.org/10.1007/s12035-020-02021-1
[4]
Kerry, R., Young, K.J., Evans, D.W., Lee, E., Georgopoulos, V., Meakins, A., et al. (2024) A Modern Way to Teach and Practice Manual Therapy. Chiropractic & Manual Therapies, 32, Article No. 17. https://doi.org/10.1186/s12998-024-00537-0
[5]
Wang, C., Guo, Z., Duan, S., He, B., Yuan, Y. and Wu, X. (2021) A Real-Time Stability Control Method through sEMG Interface for Lower Extremity Rehabilitation Exoskeletons. Frontiers in Neuroscience, 15, Article ID: 645374.
https://doi.org/10.3389/fnins.2021.645374
[6]
Wu, J., Liu, Y., Zhao, J. and Jia, Z. (2023) Research on a New Rehabilitation Robot for Balance Disorders. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 31, 3927-3936. https://doi.org/10.1109/tnsre.2023.3312692
[7]
Wang, Y. and Qiu, W.X. (2024) Inverse Kinematics Solving Method and System for Shoulder Four-Axis Rehabilitation Robot. China Invention Patent, ZL202311764361.9.
[8]
Elgammal, A.T., Magdy, M. and Lashin, M. (2024) Learning Complex Nonlinear Dynamics of a 3D Translational Parallel Manipulator Using Neural Network. Heliyon, 10, e37669. https://doi.org/10.1016/j.heliyon.2024.e37669
[9]
Mrozek, A. and Strek, T. (2022) Numerical Analysis of Dynamic Properties of an Auxetic Structure with Rotating Squares with Holes. Materials, 15, Article No. 8712.
https://doi.org/10.3390/ma15248712
[10]
Herrera, L. and Méndez-Ramírez, R. (2025) Discrete Unilateral Constrained Extended Kalman Filter in an Embedded System. Sensors, 25, Article No. 4636.
https://doi.org/10.3390/s25154636
[11]
Wang, Y. and Xu, J.Y. (2024) Motion Trajectory Prediction Method and System for Upper Limb Rehabilitation Robot. China Invention Patent, ZL202411154871.9.
[12]
Vaisali, S., Maheswari, C., Shankar, S. and Naveenkumar, R. (2025) Predicting the Performance of Assistive Device for Elderly People Using Weighted KNN Machine Learning Algorithm. Journal of Back and Musculoskeletal Rehabilitation, 38, 852-864. https://doi.org/10.1177/10538127251317602
[13]
Wang, Y. and Wei, Z.Y. (2025) Adaptive Motion Control Method and System for Wearable Upper Limb Manipulator. China Invention Patent, ZL202411464981.5.
[14]
Hocaoglu, E., Beyaz, M., Ahmed, M.A.T., Gurluk, A., Unal, R. and Kaya, P. (2025) sEMG-Based Assessment of Engagement in VR-Integrated BALANSens Ankle Rehabilitation. 2025 International Conference on Rehabilitation Robotics (ICORR), Chicago, 12-16 May 2025, 1673-1678. https://doi.org/10.1109/icorr66766.2025.11063096
[15]
Wang, Y. and Qiu, W.X. (2024) Lightweight Single-Thread Asynchronous Method and System Based on Development Environment. China Invention Patent, ZL202411117348.9.
[16]
Xu, Y., Huang, L., Zhao, T., Fang, Y. and Lin, L. (2022) A Timestamp-Independent Haptic-Visual Synchronization Method for Haptic-Based Interaction System. Sensors, 22, Article No. 5502. https://doi.org/10.3390/s22155502
[17]
Luo, Z., Wang, Y., Zhang, T. and Wang, J. (2025) Effectiveness of AI-Assisted Rehabilitation for Musculoskeletal Disorders: A Network Meta-Analysis of Pain, Range of Motion, and Functional Outcomes. Frontiers in Bioengineering and Biotechnology, 13, Article ID: 1660524. https://doi.org/10.3389/fbioe.2025.1660524
[18]
He, C., Xiong, C.H., Chen, Z.J., Fan, W., Huang, X.L. and Fu, C. (2021) Preliminary Assessment of a Postural Synergy-Based Exoskeleton for Post-Stroke Upper Limb Rehabilitation. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 29, 1795-1805. https://doi.org/10.1109/tnsre.2021.3107376
[19]
Qian, L., Song, T., Unberath, M. and Kazanzides, P. (2022) AR-Loupe: Magnified Augmented Reality by Combining an Optical See-Through Head-Mounted Display and a Loupe. IEEE Transactions on Visualization and Computer Graphics, 28, 2550-2562. https://doi.org/10.1109/tvcg.2020.3037284
[20]
Kim, J.K., Lv, Z., Park, D. and Chang, M.C. (2022) Practical Machine Learning Model to Predict the Recovery of Motor Function in Patients with Stroke. European Neurology, 85, 273-279. https://doi.org/10.1159/000522254
[21]
Stanciu, C.T., Velimirovici, M.D., Vermesan, D., Pilut, C.N., Stana, L., Bratosin, F., et al. (2025) Comparative Recovery after Acute Lower-Limb Wounds Treated with Negative-Pressure Wound Therapy and Three Gradations of Manual Rehabilitation. Healthcare, 13, Article No. 1496. https://doi.org/10.3390/healthcare13131496
