Trajectory tracking performance in task space of robot manipulators: an adaptive neural controller design

Abstract

An adaptive neural network control design for robot manipulators in task space coordinates is provided in this paper. This controller design and a direct adaptive control strategy (passivity-based controller) are simulated for the same trajectory, considering the presence of the friction torques and the influence of payload. Performances are evaluated according to behavior of position tracking, and to trajectory tracking accuracy. The adaptive neural network controller is developed based on a neural network modeling technique which neither requires the evaluation of inverse dynamical model nor the time-consuming training process, and does not require the inverse of the Jacobian matrix.