Precision Position Measurement of Linear Motors Mover Based on Temporal Image Correlation

Abstract

High-precision position measurement is very important for precision motion control of a linear motor. This paper presents an accurate mover position detection method with high precision stability based on temporal image correlation and fence image for the linear motors. First, sequence fence images with the information of mover position are recorded by a position measurement system for linear motor, and the integer pixel displacement of each image is obtained by image correlation method. Then, according to the continuity of the motion, the subpixel displacement can be extracted by fitting the integer pixel displacement. Finally, the displacement and velocity curves can be calculated according to system calibration. Compared to digital image measurement used in position detection of linear motors, the proposed method can detect mover position accurately with longer measurement range and smaller uncertainty. The moving least square (MLS) algorithm is used to ensure the stability of measurement uncertainty under different motion modes. Simulations and mover position detection experiments with different motion conditions were performed to demonstrate the effectiveness of the proposed method in linear motors. In this paper, the average measurement error of displacement is 0.006 mm with the maximum displacement of 120 mm. The maximum and average relative errors are less than 0.06% and 0.017%, respectively.