Methodology Based on Design of Experiments to Assess Quality of Estimated Disturbances by State Observers in Mechanical Systems

A methodology, based on design of experiments, to analyze the quality of estimated disturbances by state observers in mechanical systems is presented. Analysis is carried out indirectly through a control structure with active disturbances rejection, wherein the controller compensates perturbations of the plant through the estimated disturbances by an observer, if they have good quality, the error between the reference signal and the output of the plant will be small. Therefore we use two performance indices, which provide a measure of the steady-state error, to measure the quality of the estimated disturbances. The effectiveness of the methodology is illustrated by its application on a simple pendulum.