Adaptive Neural Network Fast Fractional Sliding Mode Control of a 7-DOF Exoskeleton Robot

To rehabilitate individuals with impaired upper limb (UL) functions due to neurological disorders, this research focuses on trajectory tracking control (representing passive rehabilitation exercise) of a 7 DOFs exoskeleton robot named ETS-MARSE. It is a redundant type of robotic manipulator having a very complex structure which is designed based on human UL joint articulations. The exoskeleton is constantly encountered with external disturbances and unknown dynamics such as friction forces, and backlash which is hard to model. Moreover, this type of robot needs to deal with the unknown dynamics of a wide range of subjects with different degrees of UL impairments. Therefore, to deal with this modeling uncertainty, in this paper we propose a novel adaptive neural network fast fractional integral terminal sliding mode control (ANFFITSMC) approach to maneuver the ETS-MARSE to provide passive arm movement therapy. To address the chattering phenomena which are observed in the fast fractional integral terminal sliding mode control (FFITSMC), a new adaptive radial basis function neural network (ARBFN) is incorporated with the FFITSMC. The Lyapunov theory is used in order to prove the stability of the proposed controller. Simulation results validated the efficient performance of the ANFFITSMC in terms of chattering reduction and trajectory tracking.