基于改进无迹卡尔曼滤波的弹道射程修正算法研究  

Ballistic Range Correction Algorithm Based on an Improved Unscented Kalman Filter

作　　者：雷晓云[1] 张志安[1] 杜忠华[1] LEI Xiao-yun;ZHANG Zhi-an;DU Zhong-hua(School of Mechanical Engineering,Nanjing University of Science and Technology,Nanjing 210094,Jiangsu,China)

机构地区：[1]南京理工大学机械工程学院,江苏南京210094

出　　处：《兵工学报》2018年第9期1701-1710,共10页Acta Armamentarii

基　　金：国家自然科学基金项目(11372142)

摘　　要：以全球定位系统(GPS)作为弹载测量系统的弹道射程修正弹,存在GPS动态数据易受测量噪声与系统噪声污染或在高过载环境中发生定位丢失、数据异常等问题。为了降低弹道数据测量误差,并减小对射程修正时刻预测的误差,提出了基于牛顿插值法的改进无迹卡尔曼滤波(UKF)算法,以重新估计具有粗大误差的测量数据,从而降低异常测量值和定位失锁情况对滤波效果的影响,提高UKF算法对测量数据误差的敏感程度。仿真与试验结果表明:改进的UKF算法作为预处理过程融合至射程修正算法中,当系统离散化步长与GPS数据更新周期相等时,可最大化地预处理算法效果;改进后的修正算法可有效地降低GPS数据造成的修正误差;预处理算法中数据样本容量的选择与修正时刻以及GPS更新周期相关,不受算法效果约束。For the ballistic range correction projectiles utilizing GPS as the projectile-borne detection system,the received dynamic data is susceptible to contamination from measurement noise and system noise,and data loss and exception happen in the high-overload environment.An improved unscented Kalman filter(UKF)based on Newton interpolation method is proposed to reduce the measurement error,thereby decreasing the predicted error during range correction.Data sets with abnormal errors are distinguished and reestimated.The improved UKF algorithm is used to reduce the influence of abnormal measurements and data loss on filtering effect,and more sensitive to measurement error.The improved unscented Kalman filter is designed to be adopted in an ballistic range correction algorithm.The simulated and test results indicate that the preprocessing effect could be maximized if the step size of system discretization is equal to the GPS update interval;the improved correction algorithm is able to reduce the correction errors induced by GPS data.The sample size of GPS data in preprocessing is related to the correction moment and GPS update interval,and it is not restricted by the algorithm effect.

关 键 词：弹道射程修正 预处理 无迹卡尔曼滤波 射程修正 全球定位系统 

分 类 号：TJ413.6[兵器科学与技术—火炮、自动武器与弹药工程]