2026· IEEE Transactions on Automation Science and Engineering· Vol 23, pp. 14173-14186· 0 citations· 40 references
Abstract
Lower limb exoskeletons and mobile robots hold great potential in improving motor function rehabilitation for patients with limb dysfunction. However, their widespread application is limited by the substantial time and effort investment required from professional rehabilitation therapists. A significant challenge in achieving autonomous and intelligent rehabilitation lies in addressing the coordinated control between the exoskeleton and the robotic walker. This paper proposes a novel collaborative learning control strategy based on deterministic learning, which aims to achieve high-performance coordinated control through precise closed-loop dynamics modeling of the exoskeleton-walker system. First, radial basis function neural networks (RBFNNs) are employed to approximate the system dynamics during the coordinated control process. Utilizing deterministic learning theory, it is rigorously demonstrated that, under persistent excitation conditions, the unknown dynamics of the system can be accurately approximated and stored as constant neural networks. Subsequently, an experience-based collaborative learning controller is designed, enabling autonomous coordinated control of the human-robot system and offering a viable approach for its broader application. The effectiveness and superior performance of the proposed control strategy are validated through experiments conducted on the CoppeliaSim robotic platform. Note to Practitioners—This work is intended for researchers and engineers working on rehabilitation robotics, particularly those focusing on lower limb exoskeletons and mobile robotic walkers. One of the main barriers to the practical deployment of these systems is the reliance on continuous assistance from rehabilitation professionals during use. To address this, we propose a deterministic learning-based collaborative control strategy that enables accurate modeling and reuse of system dynamics, ultimately achieving autonomous coordination between the exoskeleton and robotic walker. This reduces reliance on manual intervention and opens up the possibility for long-term, high-performance rehabilitation training in clinical and home environments. Practitioners can leverage this approach to develop smarter, more adaptive rehabilitation systems that respond to patient needs with greater precision and autonomy.
Improving trajectory-following accuracy remains a central issue in lower-limb rehabilitation exoskeletons, especially when joint motion is affected by nonlinear coupling, parameter drift, and interaction disturbances during repetitive gait training. To address this problem, this study develops an adaptive trajectory-tracking method driven by an RBF neural network. A single exoskeleton leg is first represented as a two-degree-of-freedom planar double-link mechanism in the sagittal plane, and the swing-phase dynamics are formulated via the Lagrange approach. On this basis, a nominal-model decomposition framework is introduced so that uncertain dynamics and external perturbations can be approximated online by the RBF network. A joint-space controller is then constructed to improve tracking stability and robustness. MATLAB simulations and prototype experiments are further carried out, with conventional PID control used for comparison. The results indicate that the proposed approach yields faster error attenuation, smaller steady-state oscillation, along with improved resistance to disturbances in the hip and knee motions. Overall, the proposed approach is effective for accurate gait-following control in rehabilitation exoskeletons.
Yuhui Yang, Chao Yang, Yuanxiang Guo et al.· International Conference on...· 0 citations
Lower-limb rehabilitation exoskeletons have the potential to improve gait recovery after stroke by providing intensive and repetitive training. However, conventional control strategies often rely on fixed control parameters and exhibit limited adaptability to patient-specific characteristics, sensor noise, and dynamic uncertainties. This paper proposes an adaptive control framework that combines deep reinforcement learning (RL) with model-based impedance control for personalised lower-limb exoskeleton assistance. Patient-specific biological parameters are incorporated into the simulation environment and reward formulation to improve adaptability and robustness. Three state-of-the-art deep RL algorithms, Deep Deterministic Policy Gradient (DDPG), Twin Delayed Deep Deterministic Policy Gradient (TD3), and Soft Actor-Critic (SAC), are evaluated in a continuous control environment under varying signal-to-noise ratio (SNR) conditions ranging from 5 dB to noise-free conditions. Results demonstrate that TD3 achieves the most stable learning performance, obtaining a mean reward of −354.24 under noise-free conditions, while DDPG provides the highest joint-angle tracking accuracy with an RMSE of 0.0369 rad. SAC exhibits superior robustness in noisy environments, achieving the highest learning ratio of 0.51 at 5 dB SNR. Furthermore, the proposed personalised framework reduces tracking errors by up to 27% compared with non-personalised baseline approaches. The findings indicate that integrating patient-specific information with RL-based adaptive control can significantly enhance robustness, tracking performance, and personalisation in exoskeleton-assisted gait rehabilitation, providing a promising direction for future intelligent rehabilitation systems.
Ali Foroutannia, Masoud Mohammadian, K. Munasinghe· Italian National Conference...· 0 citations
The findings indicate that integrating reinforcement learning with Sigmoid parameter adaptation provides a systematic and effective solution for adaptive compliance regulation in mobile exoskeleton systems, enhancing adaptability, safety, and functional relevance for stroke patients undergoing lower limb rehabilitation training.
In order to improve the trajectory tracking accuracy of the lower limb rehabilitation exoskeleton robot, this paper proposes an adaptive robust error compensation control method (ARCEC) based on RBF neural network. First, the dynamics of the single-leg swing phase of the lower-limb exoskeleton are modeled using Lagrange’s equations, taking into account factors such as joint friction, flexible transmission, and the torque arising from human–robot interaction. Subsequently, nominal model compensation, online approximation via RBF neural networks, and nonlinear error feedback are integrated to mitigate the impact of model uncertainty and external disturbances on trajectory tracking performance. Simulation and experimental results demonstrate that compared to traditional PID control and sliding mode control (SMC), the ARCEC method exhibits significant advantages in trajectory tracking accuracy. It achieves up to a 30% reduction in tracking error, enabling the lower-limb exoskeleton to precisely track human gait curves.
Chao Yang, Xin Han, Zhijue Huang et al.· International Conference on...· 0 citations
Results demonstrate that DRL-based methods, particularly when combined with traditional controllers, improve both force reduction and motion stability over conventional control strategies.
Mohammad Sahandi, G. Vossoughi, H. Zohoor et al.· IEEE Access· 0 citations
With the increasing aging population and the growing prevalence of upper-limb motor impairments, conventional rehabilitation approaches face limitations in terms of labor demand and long-term training continuity. This study presents a shape memory alloy (SMA)-driven upper-limb rehabilitation exoskeleton for assisting elbow and wrist joint motions. The system integrates an SMA actuator with a Bowden cable transmission mechanism to achieve a lightweight and compliant actuation design for human–robot interaction. A nonlinear SMA model based on the Liang–Rogers formulation is adopted to describe phase transformation behavior and hysteresis characteristics. A PWM-based PID control strategy was implemented to regulate joint motion and improve tracking performance under the inherent thermal delay of SMA actuation. The proposed system is evaluated through MATLAB/Simulink simulations and experimental testing. The results demonstrate that the exoskeleton can achieve stable trajectory tracking for elbow and wrist rehabilitation motions. The study confirms the feasibility of SMA-based lightweight rehabilitation systems while highlighting the influence of SMA thermal dynamics on system performance.
Sunke Zhu, Mingwei Liu, Lei Wang et al.· Engineering Research Express· 0 citations