Application of PD-type iterative learning control in hydraulically driven 6-DOF parallel platform

Abstract

The movement of the hydraulically driven six degrees of freedom (6-DOF) parallel platform is driven by six sets of valve-controlled asymmetrical cylinders (VCACs), so the precision of the platform is mainly determined by the control precision of the VCACs. Unfortunately, the asymmetrical movement of the VCACs caused by its asymmetrical structure can significantly compromise control precision because of this asymmetry, and more fundamentally, with the inherent non-linearity of hydraulic systems as well as complicated load variations, it is very difficult to achieve ideal control precision with traditional PID control. In the present study, the working principle and characteristics of a VCAC were analysed, with particular focus on the asymmetry problem. In order to improve the precision of both VCAC control and 6-DOF parallel platform movement, a PD-type iterative learning control (ILC) method is presented, and applied to a hydraulically driven 6-DOF parallel platform. Experiments on both the single VCAC system and the parallel platform were developed to verify the validity and effectiveness of this control method. The theoretical analysis and testing results, compared with those for the conventional PID control, proved that the proposed PD-type ILC method can acquire high precision on the 6-DOF parallel platform without the need to build a mathematical model or obtain accurate loading conditions.