| In recent years,MEMS gyroscope has been widely used in various fields such as aerospace,medical treatment,military and so on because of its small volume,light weight and low energy consumption.However,due to the limitation of the existing micromachining technology,the performance of domestic MEMS gyroscope can not meet the requirements of some high-precision occasions.In order to improve the performance of MEMS gyroscope,the way of laser etching can be used to modify the internal structure of the gyroscope.However,because the size of MEMS gyroscope belongs to micron level or even nanometer level,the requiement of laser spot center and processing starting point positioning accuracy is high,while the traditional positioning method with the help of microscope can not meet the requirements of positioning accuracy.In order to improve the positioning accuracy and processing efficiency,a MEMS gyroscope micro vision positioning system was developed to realize the positioning and alignment of the laser spot center and the processing starting point.The research work of this thesis mainly includes the following aspects:（1）According to the requirements of MEMS gyroscope micro vision positioning system,the selection of hardware and the construction of software environment were carried out.（2）The camera imaging model and lens distortion model were analyzed,and the camera calibration method under macro vision was investigated.According to the characteristics of micro vision,a camera calibration method based on variable magnification method was used to calibrate the micro vision system.（3）According to the gray distribution characteristics of picosecond pulse laser spot image used in laser tuning of MEMS gyroscope,a laser spot center positioning algorithm based on gray histogram was proposed.The appropriate threshold was calculated by using the gray histogram of spot image,the spot image was segmented by adaptive threshold,and then the center of target image was calculated by gray center of gravity method.The simulation and experimental results show that the positioning accuracy of this algorithm is0.309 μm under different irradiation time and 0.322 μm under different laser power,which are higher than the traditional gray center of gravity method and maximum row and column gray value method.（4）According to the microstructure of MEMS gyroscope,the vertex of triangular beam was determined as the processing starting point,and a processing starting point location algorithm based on Hough transform and least square method was proposed.Firstly,the image was preprocessed by median filter and OTSU threshold segmentation method,and then the edge pixels were extracted by Canny edge detection algorithm,Then,the combination of Hough transform method and least square method was used to fit the straight line of the edge.Finally,the vertex coordinates were calculated according to the fitted straight line equation.The simulation and experimental results show that,in terms of straight line fitting of data point set and the positioning of machining staring point of MEMS gyroscope triangular beam,the accuracy of this algorithm is higher than that of Hough transform method.（5）The experimental platform of MEMS gyroscope micro vision positioning system was built,and the human-computer interaction interface was compiled.The camera calibration algorithm,laser center and processing starting point positioning algorithm were combined with motion control to realize the positioning and alignment functions of processing starting point and laser spot center.The experimental results show that the positioning accuracy of the system in X and Y directions are 1.545 μm and 1.273 μm respectively,which meet the requirements of error 2 μm.The average positioning time is 9 s,and the efficiency is higher than that of manual positioning. |
PDF Full Text Request