Sliding-mode neural network robot controller

Abstract

The theoretical development of a trajectory tracking neural network controller based on the theory of sliding-mode controllers is shown in the paper. Derived equations of the neural network controller were verified on a direct drive 2-DOF SCARA mechanism model. The new neural network controller was compared with a neural network controller based on the computed torque method. At last we made an application on a real 2-DOF SCARA robot mechanism by the neural network controller based on the computed torque method and a sliding-mode neural network controller on a real single axis direct drive robotic manipulator driven by an induction motor.