Iterative Learning Control for a Hydraulic Cushion

Abstract

An Iterative Learning Control algorithm is presented for the force control circuit of a hydraulic cushion, which improves the existing control scheme based on classical PI control and feedback linearization. The circuit contains a proportional valve for regulating the pressure at the cylinder chamber and, therefore, the force applied by the cushion. The Iterative Learning Control filter design is based on the rejection of the valve’s dynamics, which are filtered by a fourth-order low-pass filter. The filter is divided into two second-order filters to carry out forward and backward filtering and obtain zero-phase filtering. The proposed methodology improves the performance of the existing control scheme and reduces considerably the settling time and overshoot of the pressure signal.