An active disturbance rejection controller for a parallel Robot via Generalized Proportional Integral observers

Abstract

In this article, we address an active disturbance rejection controller design for the output reference trajectory tracking problem in a 3 degree of freedom (DOF) Delta Robot. The proposed method relies on purely linear high gain disturbance observation and linear feedback control techniques. The estimation tasks are carried out with the help of Generalized Proportional Integral (GPI) observers, endowed with output integral injection to counteract zero mean measurement noise effects. As the lumped exogenous and endogenous disturbance inputs are estimated, the observers deliver them to the controllers for on-line disturbance cancelation, while simultaneously the phase variables, related to the measured flat outputs, are being estimated by the same GPI observer. The gathered values of the phase variables are used to complete a linear multivariable output feedback control scheme. The proposed control scheme avoids the traditional computed torque method, reducing the computation time and bypassing the need for explicit, accurate, knowledge of the plant. The estimation and control method is only approximate as small as desired reconstruction, or tracking, errors are guaranteed. The reported results, including laboratory experiments, are significantly better than the results provided by the classical model-based techniques, when the system is subject to endogenous and exogenous uncertainties.