摘要
针对无线定位技术应用于炮弹外弹道测量时,炮弹末段遇到定位不准确的问题,提出时变自回归(time-varying autoregressive,TVAR)时序预测炮弹弹落点的方法,根据X、Y和Z非耦合的时序点,分别建立检验最优阶数的时变自回归时序模型,使用递推最小二乘方法估计TVAR模型中的时变参数,选用傅里叶时间基,以炮弹高程与靶面相交作为判断条件,计算最后炮弹落点的靶面坐标。仿真实验选用外场无线定位弹载炮弹的末端153个采样定位数据,检验了模型外推弹落点模型预测高程落点与靶面相交时的采样时刻,估计坐标与弹坑定位测量数据坐标误差表明,时变自回归模型对炮弹落点的预测满足误差指标要求。
When applying wireless location technology to the measurement of projectile＇s outer ballistic trajectory, its terminal localization is facing an inaccurate problem. To solve this problem, this paper proposes the time-varying autoregressive （TVAR） method to model the projectile＇s outer ballistic trajectory with time and order parameters. The TVAE model for testing the optimal order was established respectively according to the non-coupling time-order points of the X, Y and Z axis, then estimate the time-varying parameter in the TVAR model by the recursive least squares method, choose the Fourier time basis, and take the intersection of artillery elevation and target surface as the condition of judgement for calculating the final target coordinates of projectile＇s point of fall. In the simulation experiment, 153 sampling location data were chosen at the end of projectile＇s outer ballistic trajectory, which is located by wireless means, and then tested the sampling moment when the estimated point of fall of the artillery elevation intesected with the target surface of the model. The error between the estimated coordinate and the measured coordinate of crater location shows that the estimation of the time-varying autoregressive model for the projectile＇s point of fall meets the requirement of error index.
作者
李黎
刘忠
刘志坤
贺静波
LI Li;LIU Zhong;LIU Zhikun;HE Jingbo(College of Electronic Engineering,Naval University of Engineering,Wuhan 430033;93 Team in 92941 Force,Huludao 125000,China)
出处
《应用科技》
CAS
2018年第4期95-99,共5页
Applied Science and Technology
基金
国家自然科学基金项目(61401493)
关键词
TVAR模型
时变系数估计
预测
炮弹落点预测
递推最小二乘估计
AIC准则
傅里叶时间基
协同定位
time-varying AR model
time-varying coefficient estimation
predict
falling prediction of ballistic trajectory
recursive least squares estimation
AIC criterion
Fourier time base
co-location
