节点文献

基于FPGA的波前零差干涉图像信号处理技术研究

Research on Wavefront Zero-difference Interference Image Signal Processing Technology Based on FPGA

【作者】 张一凡;

【导师】 于亮;

【作者基本信息】 哈尔滨工业大学 , 仪器科学与技术, 2023, 硕士

【摘要】 近年来,随着晶圆加工和超精密数控机床(工业母机)等高端装备制造行业的迅猛发展,超精密测量的需求指标加速提高。首先,位移测量需求逐渐从纳米量级延伸到亚纳米量级甚至皮米量级;另一方面,对装置测量的形式需求也从单自由度的线位移测量发展为多自由度的线位移和角度同步测量。而目前通用的多光束组合测量技术对光束平行度要求高,难以集成,单光束测量技术由于其可直接溯源,光路简洁,易于集成等特点表现出巨大的应用前景。针对目前单光束波前干涉图像测量技术存在的无法对干涉信号进行快速实时解算的问题,本文开发设计了三自由度同步解耦处理系统。主要开展了以下工作:首先,对基于波前干涉图像的零差干涉模型进行研究分析,针对自激光出射通过分光镜在相机探测器产生明暗相间的干涉条纹的光学过程进行建模,给出零差干涉的数学模型,并推导三自由度信息(位移,偏摆角和俯仰角)与空间干涉条纹的线性解耦关系,实现两个条纹像素解耦三个自由度信息。其次,研究基于三自由度解耦方法的波前干涉图像高分辨力解耦算法。通过对空间域干涉信号的像素数据分析出发,设计了从空间域到频域特征点提取的解耦算法,并进行二次曲线拟合来实现条纹像素的频域特征点高分辨力提取。同时,对设计的解耦算法进行位移和姿态调制仿真分析,结果表明角度误差小于0.2μrad,位移误差小于0.05 nm,可以满足高分辨力测量需求。然后,设计基于波前零差干涉的图像采集与解耦处理系统。由于部分像素便能够解算三自由度信息,因此基于相机ROI技术降低像素数据量,通过MIPI相机进行干涉图像的kHz量级快速采样和传输,利用FPGA进行干涉图像像素点的并行解耦处理获取三自由度信息,并通过千兆网将三自由度信息上传,实现基于波前干涉图像的三自由度信号快速在线解耦。最后,对本文所设计的测量系统进行实验测试,实验结果表明偏摆角和俯仰角分辨力达到1μrad,位移分辨力达到3 nm,其角度测量范围达到1 mrad,数据更新率达到1 kHz,满足了快速高分辨力测量需求。

【Abstract】 In recent years,with the rapid development of high-end equipment manufacturing industries such as wafer processing and ultra-precision CNC machine tools(industrial master machines),the demand indicators for ultra-precision measurement have accelerated.Firstly,the demand for displacement measurement gradually extends from nanometer scale to sub-nanometer scale or even picometer scale;on the other hand,the demand for the form of device measurement also develops from single-degree-offreedom line displacement measurement to multi-degree-of-freedom simultaneous line displacement and angle measurement.While the current general multi-beam combination measurement technology requires high beam parallelism and is difficult to integrate,single-beam measurement technology shows great application prospects due to its direct traceability,simple optical path and easy integration.To address the problem that the current single-beam wavefront interferometric measurement technique cannot perform fast real-time decoupling of the interferometric signal,this paper develops and designs a three-degree-of-freedom simultaneous decoupling processing system.The following work is mainly carried out:First,the zero-difference interference model based on wavefront interference images is studied and analyzed,and the optical process of producing bright and dark interference fringes at the camera detector from laser emission through the beam splitter is modeled,and the mathematical model of zero-difference interference is given,and the linear decoupling relationship between three degrees of freedom information(displacement,deflection angle and pitch angle)and spatial interference fringes is derived to realize two fringe pixels to decouple three degrees of freedom information.Secondly,the wavefront interferometric image high-resolution decoupling algorithm based on the three-degree-of-freedom decoupling method is studied.The decoupling algorithm from spatial domain to frequency domain feature point extraction is designed by analyzing the pixel data of the spatial domain interference signal,and the quadratic curve fitting is performed to realize the frequency domain feature point extraction of stripe pixels with high resolution.Meanwhile,the designed decoupling algorithm is simulated and analyzed for displacement and attitude modulation,and the results show that the angular error is less than 0.2 μrad and the displacement error is less than 0.05 nm,which can meet the demand of high-resolution measurement.Then,a wavefront zero-difference interference-based image acquisition and decoupling processing system is designed.Since some pixels can decode the threedegree-of-freedom information,the amount of pixel data is reduced based on the ROI(region of interest)technology,and the MIPI camera is used to sample and transmit the interferometric image at kHz level quickly,and the FPGA is used to decouple the pixel points of the interferometric image in parallel to obtain the three-degree-of-freedom information,and the three-degree-of-freedom information is uploaded through the gigabit network.The three-degree-of-freedom information is uploaded through the gigabit network to realize the fast online decoupling of three-degree-of-freedom signals based on wavefront interferometric images.Finally,the experimental test of the measurement system designed in this paper shows that the resolution of deflection and pitch angles reaches 1 μrad,the resolution of displacement reaches 3 nm,its angle measurement range reaches 1 mrad,and the data update rate reaches 1 kHz,which meets the demand of fast and high resolution measurement.

  • 【分类号】TH744.3;TP391.41
节点文献中: 

本文链接的文献网络图示:

本文的引文网络