Robust high precision multi-frame motion detection for PMLSMs’ mover based on local upsampling moving least square method

Abstract

This paper studies a multi-frame visual sub-pixel measurement technique to achieve the precise positioning of the permanent magnet linear synchronous motor (PMLSM) rapidly. A new matrix multiplication discrete Fourier transform (DFT) upsampling moving least square (MLS) algorithm is proposed to improve the accuracy and robustness of PMLSM positioning. Firstly, a line-scan vision-based measurement system is designed, and an aperiodic sinusoidal fringe image is placed on the pedestal as target image. Secondly, the local upsampling cross-correlation curve of adjacent one-dimensional signals is calculated to obtain sub-pixel displacement using the matrix-multiplied DFT method. Finally, to further improve the displacement accuracy, the MLS method is applied. Simulation tests verify that the proposed method is superior to other algorithms in measurement accuracy, calculation efficiency and measurement range. The experiment results confirm the feasibility and validity of the proposed method for detecting the mover motion under different working environments.