Research And Implementation Of Single-ring Absolute Encoder Subdivision Technique

Single-ring absolute photoelectric shaft encoder uses single channel coding technology and linear CCD image measurement and subdivision technology,which also processes advantages of high resolution,small,high precision and non-contact and so on.The different width characterizations of stripes depicted on the single-ring absolute disc represent different coding numbers,and the centers of stripes divide the disc circumference into same areas.The coding value can be obtained from the combination of 12 bits coding numbers.The basic working principle of the encoder is using light-emitting diode(LED)as a light source to irradiate the coding disc,and then the stripe image of coding disc can be received through optical signal receiving device by using linear array CCD.CCD transforms the spatial light intensity signal of the position information into electrical signals,and then the digital signal is achieved by AD sampling and then sent into FPGA to accomplish decoding and subdivision processing.On one hand the rough angle degree is achieved by comparing the decoding value with the absolute angle information which is store in ROM after decoding by the code value recognition module.On the other hand the precise location of the stripes is obtained by using CCD subdivision technique.Finally,the value of angle is precisely received from combination of these two values.Single-ring absolute encoder code recognition and subdivision are implemented based on FPGA in this paper by researching the encoding and decoding principle and linear array CCD subdivision principle and the measurement precision reaches 2 ".Firstly,this thesis detailed analyses the single-ring absolute encoder's coding principle and its concrete realization.Secondly,photoelectric detection and image acquisition system are designed based on FPGA and linear CCD.Thirdly the single-ring absolute encoder decoding module is implemented based on FPGA,including threshold automatic calculation module,image binary module,code value recognition module,rough measurement value calculation module.The applicability of the polynomial interpolation algorithm and centroid location algorithm is analyzed through simulation.Polynomial interpolation algorithm is not suitable for the narrow stripe with small number of effective pixels.Edge orientation can only get single edge location information which also causes wrong result.Discontinuous change of centroid position occurs when the pixels which participate in operation are changed.Centroid location jumps 0.4 pixels by simulation analysis and measurement proves the conclusion.The centroid calculation with threshold algorithm ensures the centroid location continuous change which also gets the correct result.The precise value is obtained by calculating the shifting between the centroid of image and virtual center.The centroid of stripe is calculated based on CCD centroid location algorithm.Finally,this thesis measures the precision of different subdivision algorithm,and summarizes the influence factors and solutions of the error in the realization process.The conclusions provide the experience and theoretical basis for the following project.