Research On Angle Measurement Technology Of Absolute Circular Grating Based On Image

As a high-precision non-contact angle measuring instrument,the circular grating photoelectric encoder has many advantages,such as wide measuring range,good reliability,high resolution,non-contact wear and so on.It is widely used in the dynamic measurement and real-time control system of national defense,military,aerospace and other scientific and technological fields.However,the traditional photoelectric encoder has the problems of complex signal processing,difficult marking and installation process,and it usually improves the resolution by increasing the size of the code disk,which cannot meet the strict requirements for the volume and weight of devices in the rapidly developing precision measurement and automatic control fields.In this paper,aiming at the shortcomings of traditional encoder,the angle measurement technology of image single-ring absolute encoder is studied from the aspects of coding mode,coarse code recognition and fine code subdivision.In the exploration of coding principle,aiming at the problem that multichannel coding is not easy to be miniaturized,the Huffman coding algorithm is proposed to realize the single circle grating coding method,the 16 bit single sequence-shifting code coding is realized,and the 16 bit decoding table is established,which lays a foundation for the decoding work;The angle measurement process is more intuitive by using image processing technology,and the measurement resolution can be improved based on the image pixel or sub-pixel size.In the coarse code recognition,based on the one-dimensional bar code recognition technology,a complete code disk bar code recognition algorithm is given,and the commonly used width measurement method in bar code recognition is improved.At the same time,a similarity matching algorithm is proposed to correct the recognition results and complete the 16 bit code disk bar code recognition;In order to improve the angle measurement accuracy of the encoder,the image processing technology of binomial interpolation is used to locate the barcode edge to realize the precise code subdivision,and the final angle measurement value is obtained by combining the coarse code with the precise code.In order to verify the effectiveness of coarse code recognition and similarity matching correction algorithm,an experimental platform was built to collect raster image data,the experimental results show that the recognition accuracy of the traditional algorithm is 63.80%,the improved algorithm is 90.18%,the recognition accuracy is 26.38%higher than the traditional algorithm,and the similarity matching method corrects the wrong coding sequence,so that all the coarse codes are recognized correctly.Secondly,to verify the validity the validity of the complication method and image segmentation,16 bit experimental grating is made in the encoding mode of this paper,and the measured value of Renishaw grating product is taken as the true value.The error between the measured value of experimental grating and the true value is compared,and the error data is analyzed,The experimental results show that the angular resolution of the image-type single-turn photoelectric encoder designed using the coding and subdivision research methods in this paper can reach about 2²²bit,and the angular accuracy is about 1.7193".Finally,angle measurement accuracy of circular grating is affected by systematic error are explored,which include the the grating lines positional error and the subdivision error.And these two errors are related to coding mode and subdivision principle.The analysis of the spectrum characteristics shows that the multi reading head average method has a good compensation effect on the high frequency systematic error,and greatly improves the angle measurement accuracy of the circular grating encoder.