2026· IEEE Transactions on Automation Science and Engineering· Vol 23, pp. 12486-12498· 0 citations· 34 references
Computer Science
Abstract
Assist-as-needed (AAN) assistance can encourage active participation during human walking. However, individuals exhibit diverse walking patterns, which makes it challenging for exoskeleton to provide personalized assistance. This paper presents an innovative adaptive AAN control strategy with hybrid torque fusion, including two modules of hybrid torque fusion estimation and adaptive control to promote voluntary participation. Specifically, the hybrid torque fusion module fuses torque estimated from surface electromyography (sEMG) signals with dynamic model estimation using regularized particle filter fusion (RPFF) algorithm to evaluate the human’s active participation. The control module incorporates a globally continuous extended assistance level function (EALF) that integrates joint motion tracking error, human-robot interaction force, and voluntary deficit to quantify subject performance and enable smooth transitions between four training modes, thereby continuously adjusts the torque output. Comprehensive comparisons with existing AAN strategies under multiple scenarios demonstrate that the proposed method improves trajectory tracking accuracy by approximately 16.1% and reduces human-robot interaction force by 4.01% during slope walking and simulated gait impairment. NASA-TLX and Likert scale assessments further validate the method’s effectiveness in enhancing active participation and wearing comfort. Note to Practitioners—To address insufficient active participation and unsmooth assistance in walking rehabilitation using ankle exoskeleton, caused by incomplete quantification of human motor ability, this study introduces a human active torque fusion mechanism. We developed a personalized human-robot collaboration model and proposed an AAN control algorithm based on an assistance function. This algorithm can potentially be transferred or extended to other rehabilitation robotic platforms. Although the system demonstrates promising application potential, this study acknowledges certain limitations, such as a small sample size and insufficient exploration of diverse patient population needs. Future research can build on this work by integrating additional sensing technologies, optimizing control strategies to enhance system adaptability, and exploring more efficient adaptive or online learning methods to simplify the parameter calibration process before training. This study offers the potential to significantly improve daily living for individuals with mobility limitations, laying the groundwork for more personalized and effective rehabilitation solutions.
In this paper, we propose an assist-as-needed (AAN) backstepping control scheme for a lower-limb exoskeleton with nonlinear dynamics and uncertain human–robot interactions. The main objective is to achieve a good trajectory tracking capability while adaptively controlling the assistance of the robot according to the user’s effort. The adopted dynamic model is nonlinear, which includes joint dynamics and external human interaction torque. This allows for the derivation of the tracking error formulation. The backstepping control law, formulated based on the filtered tracking error, ensures stable closed-loop performance with bounded tracking errors. We incorporate an AAN scaling framework based on estimated human effort to regulate the overall control torque as a convex combination of the nominal backstepping torque and the impedance-based assistance torque. The proposed controller was tested by numerical simulations and was compared with the sliding mode control (SMC) and the proportional–integral–derivative (PID) control. The overall root-mean-square tracking error for the proposed controller was 0.0962 rad, while for the SMC controller and PID controller, it was 0.0819 rad and 0.1246 rad, respectively. Moreover, the proposed controller reduced the peak human–robot interaction torque to 14.68 N·m compared to 15.36 N·m for SMC and 15.81 N·m for PID, adaptively controlling assistance based on the applied effort of the user. The assistance ratio went down from an average of 0.7988 in the low-effort condition to 0.6960 in the higher-effort condition, indicating effective adaptation while maintaining stable tracking performance. Although the PID controller achieved the lowest torque-variation index, the proposed controller achieved a more favorable trade-off among tracking accuracy, adaptive assistance, and acceptable torque smoothness. Finally, the proposed AAN backstepping controller achieved a practical trade-off between tracking accuracy, adaptive assistance, torque smoothness, and interaction safety, suggesting its potential in rehabilitation and assistive exoskeleton applications.
Muktar Fatihu Hamza, A. I. Isa, Abdulrahman Alqahtani et al.· Applied Sciences· 0 citations
Lower limb exoskeletons are designed to assist dependent individuals in their daily activities, such as walking, sitting, or standing, and are also intended for use as devices to support neuromuscular rehabilitation. Among lower limb exoskeletons, Active Ankle-Foot Orthoses (AAFOs) show great potential for improving user mobility. However, accurately developing a realistic dynamic model of the AAFO-wearer system for effective control remains a challenging task. This paper proposes a novel finite-time output-feedback non-model-based adaptive control approach for an AAFO. The approach uses two neural networks (NNs): the first one serves as an adaptive controller for position tracking, while the second is used to estimate external torques acting on the AAFO-wearer system. The adaptive law of the first NN relies on the torque error estimated by a Fuzzy Inference System (FIS), while that of the second NN is based on the tracking error. The proposed approach is designed to provide adaptive assistance during walking based on the estimated external torque. The proposed approach does not require any knowledge of the AAFO-wearer system dynamics and is able to adapt to different walking speeds. The performance of the proposed approach, in terms of trajectory tracking accuracy and robustness against external disturbances and parameter uncertainties, was evaluated through simulations and real-time experiments involving five healthy subjects, and then compared to that of baseline controllers from the literature. Note to Practitioners—This paper proposes a robust non-model-based finite-time output-feedback adaptive control approach for an AAFO intended to assist plantarflexion and dorsiflexion movements of the ankle joint in the sagittal plane. Achieving reliable assistance is challenging because the AAFO-wearer dynamics is affected by uncertainties, unmodeled human-robot interactions, and external disturbances such as ground-reaction forces. The approach uses two neural networks (NNs): the first serves as an adaptive controller for position tracking, while the second estimates the external torques acting on the AAFO-wearer system. The adaptive law of the first NN relies on the torque error estimated by a Fuzzy Inference System (FIS), while that of the second NN is based on the tracking error. The proposed approach is designed to adapt to different gait patterns and maintain both tracking accuracy and robustness with respect to uncertainties and external disturbances. It is validated through simulations and real-time experiments, showing improved trajectory tracking and reduced muscular effort. Future work will focus on validation with real patients under different walking environments.
Oussama Bey, M. Chemachema, R. Jradi et al.· IEEE Transactions on Automat...· 0 citations
Providing assistance across diverse movements is a central objective of exoskeletons, and anatomical knowledge can enable responsive support that generalizes across tasks. However, anatomical assistance has mainly been studied for lower-limb exoskeletons, where periodic, weight-bearing motions impose lower demands on torque precision. Extending such assistance to complex, nonperiodic upper-limb movements remains challenging. This paper proposes Anatomical Torque with Passivity-Based Control (ATP) for safe upper-limb exoskeleton assistance. First, a scalable musculoskeletal simulation framework trains a unified reinforcement-learning muscle controller that generalizes across upper-limb movements and generates anatomical reference torques without complex biomechanical computations. Second, an online torque-refinement scheme adapts the reference to diverse movements, suppresses tendon-induced spikes, and incorporates a learned anomaly score for safe and comfortable assistance. Third, an interaction torque controller delivers assistance through a cable-driven compliant exoskeleton without constraining motion to predefined trajectories, while an energy tank preserves passivity with theoretical guarantees on torque tracking and system passivity. Simulations and real-world experiments show accurate tracking of long-duration motion sequences and generalization to real-time human movements. The controller achieves accurate torque tracking while preserving passivity and resumes tracking after energy-tank replenishment. An EMG study with five participants further shows reduced target-muscle activity during static and dynamic tasks compared with gravity compensation and open-loop assistance, with reductions of up to 48% relative to movement without the exoskeleton in a dynamic multi-joint task.
This study presents a comprehensive sim-to-real approach aimed at determining the optimal assistance torque for hip exoskeletons, based on reinforcement learning and musculoskeletal simulation. A coupled human-exoskeleton modeling framework was developed, in which a bilateral hip exoskeleton with dynamic properties was explicitly integrated into an anatomically-accurate human musculoskeletal model. Reinforcement learning (RL) algorithm was employed with a phase-aware policy trained utilizing stage-wise learning, in order to generate a bilateral hip assistive torque profile. The learned torque profiles exhibited a stable and consistent biphasic pattern over the gait cycle. The musculoskeletal simulations showed that the gluteus maximus and the flexor digitorum longus muscles had reduced activation at their functional phases with exoskeleton assistance. The learned torque profile was then directly deployed to a customized hip exoskeleton system for real-word applications. In the preliminary metabolic experiments, the overall metabolic power decreased by 3.8%, and baseline-subtracted power decreased by 6.1% relative to the walking without exoskeleton condition. These findings offer compelling evidence that underscores the pivotal role of integrating sim-to-real techniques with musculoskeletal modeling in enhancing walking-assistive exoskeletons, paving the way for more advanced control strategies and bridging the gap between simulation and real-world deployment.
Fanjie Wang, Jiaxing Zhang, Ji Huang et al.· 2026 IEEE International Conf...· 0 citations
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.
Weitian He, Xinhao Zhang, Qinchen Yang et al.· IEEE Transactions on Automat...· 0 citations
The ankle joint is the core hub of human walking and bears the gait load. Its flexibility and stability are crucial to balance walking efficiency, and damage prevention. Insufficient dorsiflexion torque can significantly reduce walking efficiency and stability. Therefore, this paper designs an ankle exoskeleton to assist dorsiflexion motion. Firstly, the system adopts a rigid-flexible integrated structural design, meeting the wearer’s rigid support while reducing the quality of the exoskeleton system and increasing the flexibility to the wearer. Meanwhile, a cable-driven and load alignment transmission structure was designed to improve interaction comfort and data acquisition accuracy. Secondly, a multi-sensor feature-level perception system fused with hidden markov models (HMM) is proposed to accurately identify different gait types. Finally, an impedance control strategy based on force-position feedback is constructed to provide suitable assistive moments for users to walk at different velocities. Four subjects participated in surface electromyography (sEMG) and metabolic cost experiments. Experimental results show that exoskeleton equipment achieves precise assisting ranges, and the Rate of Change with average dorsiflexion sEMG signal and metabolic cost reduction with exoskeleton assistance is 18.52% and 11.49% respectively compared with unwearing conditions.
Yuqin Xu, Shisheng Zhang, Jinshi Zhang et al.· IEEE Transactions on Medical...· 0 citations
