Multirate Adaptive Robust Control with Friction Estimation and Compensation for Tilting Table Machine Tools

This paper proposes an advanced position control system for the 5-axis machine tool with a tilting table. The challenge is to simultaneously deal with two types of nonlinearity: the position-dependent nonlinearity of unbalanced torque, and the velocity-dependent nonlinearity of Coulomb friction. To overcome this issue, this paper develops a double-layer parameter identification (DLPID) based on the conditional recursive-least square algorithm. The DLPID provides the real-time estimation of not only the inertia, damping, and the maximum unbalance torque but also the friction coefficient. We established a two-degree-of-freedom control configuration in which the multi-rate feedforward controller and the multirate nonlinearity compensator (MNC) are adaptively updated by the DLPID. Experiment results show that the proposed control system can reduce the root-mean-square deviation of the position tracking by up to 62.3% in comparison with a conventional PID controller.