Research On Fiber Optic Winding Machine Cable System Based On Image Monitoring

In inertial navigation system,gyroscope as a new type of sensitive components,provide posture,angular velocity and other important information for the inertial navigation system.At present,it is widely used in aerospace,navigation and defense science and technology and other important areas of the country and has important strategic significance.Aiming at the problems such as poor precision,poor control system,low winding efficiency and consistency in the process of winding fiber optic gyro,this paper proposed a method to improve the processing technology by machine vision assistant detection.The main contents of this paper include the following:(1)The mechanical structure of the traditional optical fiber winding machine and its mutual movement were analyzed.In order to replace the manual line,dial line institutions were designed.Compared with the advantages and disadvantages of a variety of fiber winding process,the four-pole symmetrical winding process of the traditional filament winding machine was analyzed in detail,and the winding process based on image monitoring was completed.(2)The selection of the camera,lens and light source in the visual monitoring system of the fiber winding machine was completed.The characteristics of the installation method of various image acquisition systems were analyzed,and it was determined that Ming field diffuse backlighting is the most suitable installation scheme of the system.(3)The algorithm for obtaining the region of interest of the fiber dial wheel was designed to solve the drawback that the camera has captured a large amount of image data and contains a large amount of unnecessary information.A variety of fiber contour image enhancement,image segmentation and edge detection algorithms were compared.Finally,the basic image processing was carried out by image segmentation with median filter and dynamic iteration threshold,and then canny edge detection.(4)An algorithm for obtaining the peak position of the fiber from the region of interest of the fiber profile was designed.In this paper,the position stability of the wave crest was improved by the interpolation and filtering method for the uncertainties,discontinuities and the error peak position of the fiber contours.Then,according to the location law of the crest,the image processing and monitoring algorithm of the gap fault and the overlapped fault were designed.(5)Based on the analysis of the peak position information obtained in(4),a linear fitting prediction algorithm for predicting the position of the fiber was designed.The experimental results show that the prediction algorithm is effective.By predicting the position of the fiber,the pulley wheel movement increment was obtained,and then the final fiber winding machine's automatic dial line control was completed.(6)The multi-axis motion controller based on ARM and FPGA was designed to replace the original PLC control scheme and improve the flexibility of the control system.Experiments show that the multi-axis synchronous controller has a good synchronization control effect.