Adaptive Robust Impedance Control and Neural Network Based Control for Telerehabilitation with Upper-Limb Robotic Exoskeletons
With the unprecedented increase in global senior population and the accompanying debilitating neurological diseases, there is a growing and unmet demand for physical rehabilitation. Additionally, the drastic reduction of in-person care and the increased physical and cerebrovascular injuries during the current pandemic is not only interrupting the essential continuum of physiotherapy, but also causing the divergence between the demand and the supply of the service to grow even more. Telerehabilitation with robotic exoskeletons is an emerging, and compelling complementary rehabilitation modality which could help address the widening gap between the demand and the supply of physiotherapeutic services. Some of the prevailing challenges for the robot controllers are to overcome the effects of dynamic modeling uncertainties and ensure good tracking performance, stability, safe and compliant motion, and a high degree of telepresence between the two remotely-separated human-robot systems in the presence of nonlinearities, human torques, and communication constraints such as time delays. To address these challenges, two control methods are further developed, implemented, and validated for telerehabilitation with upper-limb robotic exoskeletons: Adaptive Robust Integral Impedance model (ARII) control and Adaptive Robust Integral Radial Basis Function Neural Networks-based Impedance model (RBFNN-I) control. Both implementations have been extended to provide compliant behaviour using an adjustable impedance model controller and revealed desirable performance for the conditions used in this research. A salient contribution of this research is the creation and implementation of a novel human torque regulator (HTR) which was shown to provide higher fidelity telepresence for the therapist compared to existing methods to enhance the safety and perception of the closed-loop physical interaction. Experiments were performed using single-joint robots while simulations were carried out using two-degrees-of-freedom (2-DOF) exoskeletons models to validate the proposed controllers and advance the state-of-the-art control for telerehabilitation with upper-limb robotic exoskeletons.