Experimental validation of control for a positioning stage by feedback error learning

This paper considers precision positioning of a stage. To compensate uncertainties of the stage, feedback error learning (FEL) is employed. In FEL algorithm, parameters of feedback error learning are updated by using an output of a feedback controller. After learning, the feedforward controller with an inverse system of a plant is obtained. To shorten learning time, implementation methods of FEL are proposed. In our methods, free parameters in the feedforward controller are determined by using the inverse system. Furthermore, information on the inverse system is utilized for the learning. Proposed methods are demonstrated by experiments with the stage.