Nonlinear Friction Modeling and Compensation for Precision Control of a Mechanical Feed-Drive System

Friction exists in mechanical systems, such as machine-tool feed drives, and causes undesired position tracking errors. The most difficult problem in friction compensation is nonlinear friction during changes in motion direction or in low velocity regions causing significant tracking errors. Many static and dynamic friction models have been proposed to compensate for frictional effects to reduce the tracking error in the desired trajectory, in particular, in low velocity regions. Although recently developed dynamic friction models consider the pre-sliding regime friction behavior and friction properties in regions of near-zero velocity, these existing friction models consider only limited sources of friction. In this paper, we present a new friction model that considers many friction sources with complicated and nonlinear properties not only in near-zero velocity regions but also in higher velocity regions. In addition, we present a controller design that includes a feed forward compensation term with the proposed friction model and a disturbance observer. Experiments were conducted to compare the control performance between the proposed and conventional friction models. The proposed controller largely improved the control performance, reducing the maximum contouring error to less than 1.6 mu m.