Investigation on Tracking Performance of Adaptive Friction Compensation Using Cascade P/PI Controller at Low Velocity

Tracking performance of drive systems is one of the factors that contribute to the accuracy of a machine tool. Tracking error which is much related to non-linear friction behavior of the system, is analyzed based on sliding and pre-sliding regime in terms of quadrant glitches that occurred in reversal velocity. In pre-sliding regime, friction forces is described while axes moving at low velocity. Friction compensation model is therefore a need for accurate motion control applications. This paper analyzes experimentally the performance of two different friction compensation model based namely Generalized Maxwell Slip (GMS) and Sigmoid like curve function (SLCF) model. The experiment validation is performed in a circular motion tested at a ball screw driven XY table controlled by cascade P/PI controller feedforward. The experimental results indicates that SLCF friction model based feedforward capable to reduce magnitude of quadrant glitch that lead towards better tracking performance in machine tools application.