Absolute optical imaging position encoder

Abstract

An absolute encoder is widely employed in a computerized numerical control machine. In order to find a zero position, an incremental encoder must have a limit switch and often return to zero. But an absolute encoder can determine the absolute position of each point. So, it has high efficiency and its structure is simpler compared to the incremental encoder. However, the accuracy and the price of absolute encoders remain to be improved. To improve the measurement accuracy, a novel absolute optical imaging position encoder is proposed. The proposed encoder is based on the pseudo-random sequence, and a CMOS sensor is applied for collecting coding pattern images. A DSP system performs decoding task. A special image processing method is used to improve measurement accuracy. The measurement errors of the images are analyzed. Glass grating is processed by the traditional technology with line width of 10μm, so the manufacturing cost is decreased accordingly. Experimental results show that the proposed image processing method can improve the measurement accuracy of the proposed absolute optical imaging position encoder by several microns.