Intelligent nonsingular terminal sliding mode control for single-degree-of-freedom knee rehabilitation system

In this paper, an intelligent non-singular terminal sliding mode (INTSMC) controller is proposed to control the single-degree-of-freedom knee rehabilitation system. This continuous passive motion (CPM) machine base on constraint-induced movement therapy (CIMT) is for patients who are receiving knee rehabilitation. First we use principle of virtual work to obtain the dynamic equation of CPM machine. An INTSMC which involves recurrent Chebyshev neural network (RCNN) estimator to estimate and compensate the unknown matching disturbance and the uncertainty of CPM machine online is proposed to achieve finite time tracking control. Finally some experimental results are obtained to show the validity of the proposed control approach.