Model Predictive Control for Upper Limb Rehabilitation Robotic System Under Noisy Condition

Abstract

Demands for rehabilitation robots are now increasing day by day due to increase in the number of patients with neural disorder. These robots help the patients in therapeutic exercise performing specific movements which leads to mitigating neural disorders through a gradual improvement of the patients' limb performances. As robots are the best suitable options to perform repetitive tasks without the risks of monotony and fatigue failure, rehabilitation via robots have proven to be more of a comfortable exercise than an exhausting treatment procedure. Rehabilitation robots require precise and efficient control in terms of position and force, ensuring thus accuracy in exercise movements, ensuring with element of enjoyment patients' safety. Nonlinear controllers make good option to this end as they adapt to handling the system uncertainties and parametric changes. This paper presents a Model Predictive Control (MPC) to control the rehabilitation robot for upper limb extremity under disturbed conditions. From results maximum overshoot of 1.4 and 1.0 and steady state error of 0.99 is found under disturbed and noisy condition respectively. Hence MPC proves to be a robust controller of external disturbances rejection and noise filtration.