一种利用PCA的导弹轨迹重建方法  

A Missile Trajectory Reconstruction Method Based on PCA

作　　者：左峥嵘[1] 吴婷[1] 

机构地区：[1]华中科技大学自动化学院多谱信息处理国防重点实验室,武汉430074

出　　处：《宇航学报》2013年第11期1468-1474,共7页Journal of Astronautics

基　　金：航天创新基金项目(CASC201104)

摘　　要：非同时刻成像将大幅恶化立体视觉方法对导弹的轨迹点重建精度和射向估计精度,针对这一问题,该文提出了应用主成分分析(PCA)重建导弹轨迹并且估计导弹射向的方法。论文在同步轨道双星观测条件下推导了将PCA方法应用到导弹轨迹重建及射向估计的原理,论证了目标三维轨迹最小方差投影直线在像面上的投影即为目标成像轨迹的最小方差投影直线,给出了通过求取目标成像轨迹的最小方差投影直线重建导弹发射面和轨迹点的算法。仿真实验表明了该方法可以有效地进行弹道轨迹的三维重建及射向估计,与现有方法相比,PCA方法重建精度更高,相机定标误差在一定范围内时,射向估计误差更小。Asynchronous imaging of dual cameras usually deteriorates the estimation of ballistic trajectory. To improve the performance of the estimation, a method for reconstructing missile trajectory and estimating missile course based on principal component analysis （PCA） is proposed. Firstly, the principle of missile trajectory reconstruction and course estimation based on PCA is derived under the background of binary observation on geosynchronous orbit. It is proved that the PCA transformed image trajectory of target is exactly the projection of the PCA transformed 3D trajectory. Then, the algorithm for estimating the missile course and trajectory through the PCA transformed image trajectory is proposed, and the reconstruction experiments are used to verify the proposed PCA method＇ s feasibility and availability for 3D reconstruction and course estimation of missile trajectory, Compared with the traditional stereo-vision based method, the PCA method has higher trajectory reconstruction accuracy and smaller course estimation error within a certain range of camera calibration error.

关 键 词：三维重建 射向估计 主成分分析 导弹轨迹 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]