Tracking performance characterization and improvement of a piezoactuated micropositioning system based on an empirical index

Motion tracking is an important problem in micropositioning systems dedicated to ultra-precision robotic micromanipulation. This paper investigates the periodic motion tracking performance of a micropositioning system based on an empirical tracking performance index (TPI), which is proposed with the consideration of both tracking errors and tracking frequencies. To improve the TPI of a micropositioning system with piezoelectric actuation, a two-loop controller consisting of an H∞ robust controller and a disturbance observer is implemented. Experimental studies are conducted on an XY parallel micropositioning system for the motion tracking tests. Results demonstrate that a sole H∞ robust controller is insufficient to produce a satisfactory tracking performance with low-frequency control. In contrast, the two-loop controller endows the system with both a lower tracking error and a better TPI than the conventional robust control does. The experimental results not only validate the effectiveness of the proposed control method in improving tracking performance but also confirm the feasibility of the TPI in quantitatively characterizing the performance of a micropositioning system.