Mover Position Detection for PMSLM Based on Line-Scanning Fence Pattern and Subpixel Polynomial Fitting Algorithm

Abstract

The article presents a rapid and precision position detection method for permanent magnet synchronous linear motors (PMSLM) based on digital image measurement and line-scanning photography. An exclusive fence pattern is designed as the target image for image measurement method, according to the motion feature of PMSLM. A line scan camera is installed on the mover to record image sequences instantaneously with the movement of linear motor, and from which the pixel displacement can be obtained using an image matching algorithm. To further improve the measurement accuracy, a subpixel polynomial fitting algorithm is proposed based on the line-scanning fence patterns. Finally, the displacement can be calculated according to system magnification parameters. Compared with the traditional image measurement method based on the area-array speckle pattern, both the cost and time delay of the proposed method are reduced greatly by employing line-scanning photography. The detection accuracy is also improved by four times. To improve the measurement robustness, an assessment index of fence pattern is proposed to optimize the fence image. The measurement error and robustness are analyzed in theory and simulation, and it demonstrates that the fence pattern shows stronger resistance to scanning deviation than speckle pattern. A mover position detection platform is established, and the experimental results show that the proposed method can achieve position estimation with average error of 0.005 mm under different working conditions.