基于粒子群算法的旋转双棱镜指向误差校正方法  

Pointing Error Correction of Risley-Prism System Based on Particle Swarm Algorithm

作　　者：沈英[1] 黎乐谦 黄峰 任和 刘锦春 Shen Ying;Li Leqian;Huang Feng;Ren He;Liu Jinchun(School of Mechanical Engineering and Automation,Fuzhou University,Fuzhou,Fujian 350108,China)

机构地区：[1]福州大学机械工程及自动化学院,福建福州350108

出　　处：《光学学报》2021年第24期131-140,共10页Acta Optica Sinica

基　　金：国家自然科学基金(62005049);福建省自然科学基金(2020J01451);福建省青年基金(2020J05101);福建省教育厅青年基金(JAT200032);福州大学科研启动基金(GXRC-20049)。

摘　　要：旋转双棱镜系统具有扩大视场的特点,在大视场、高精度目标跟踪领域具有广泛的应用前景。通过增大棱镜顶角可以提高视场放大倍率,但同时增大了系统装配误差对目标指向精度的影响。针对大顶角旋转双棱镜系统指向精度不高的问题,提出一种基于粒子群算法的旋转双棱镜指向误差校正方法,搭建了棱镜顶角为14.85°、棱镜折射率为1.515的旋转双棱镜成像系统;基于理想模型的指向测试结果,建立了基于装配误差分析的实验样机数学模型,实现对反向求解算法中棱镜顶角设定值和折射率设定值的参数辨识,最后在实验样机中进行指向测试。实验结果表明该方法可以有效提高大顶角棱镜的旋转双棱镜系统的指向精度,校正前后的最大指向误差减小了52.4%,平均指向误差减小了43.3%,均方根误差减小了44.0%,最小二乘法拟合圆半径减小了44.7%。The Risley-prism system has a great application prospect in target tracking with the aim of both large field of view and high precision due to its extension of the field of view.Although increasing the prism vertex angle enlarges the magnification of the field of view,it also expands the influence of the system assembly error on the target pointing accuracy.Considering the low pointing accuracy of the Risley-prism system with a large vertex angle,this paper proposes a method of correcting the pointing error of the Risley-prism system based on the particle swarm algorithm.A Risley-prism imaging system is built with a prism vertex angle of 14.85° and a prism refractive index of 1.515.A mathematical model of an experimental prototype based on assembly error analysis is built by using pointing test results of the ideal model.Parameter identification of the setting values of the prism vertex angle and the prism refractive index in the reverse solution algorithm is conducted.Finally,a pointing test is carried out on the experimental prototype.The experimental results demonstrate that the proposed method effectively improves the pointing accuracy of the Risley-prism system with a large prism vertex angle.When the method is applied,the maximum pointing error decreases by 52.4%,and the average pointing error reduces by 43.3%.The root-mean-square error decreases by 44.0%,and the fitting radius of the least squares method decreases by 44.7%.

关 键 词：测量 棱镜 旋转双棱镜 大顶角棱镜 粒子群算法 指向误差校正 

分 类 号：O436[机械工程—光学工程]