一种基于多模型算法的纯弹道式弹道落点预报方法

An IMM-Based Impact Point Prediction Method of Ballistic Target

本文以落点预报技术在弹道目标预警中的应用为背景,主要解决在未知弹道系数情况下如何进行弹道目标落点精确预报的问题。为此,本文给出了一种基于交互式多模型算法(IMM)和无迹滤波(UF)算法的精确弹道目标落点预报方法,从弹道系数估计收敛速度和最终收敛精度的角度对该方法落点预报的时效性和精度进行了分析,同时还分析了雷达数据率对算法性能的影响。与基于单一的UF算法或推广卡尔曼(EKF)滤波算法的落点预报方法相比较,基于IMM-UF的落点预报方法具有更强的适应性,能在未知弹道系数的情况下更迅速且更精确地估计出目标的弹道系数,从而实现对弹道目标落点更快更准的预报。仿真结果验证了该算法的有效性。

For the application of impact point predicting technique in ballistic targets early warning,a method for predicting impact point of ballistic target with unknown ballistic coefficients based on the combination of the Interacting Multiple-Model （IMM） and the Unscented Filter （UF） algorithm is presented. Timeliness and accuracy of the impact point prediction are analyzed from aspects of the convergent speed and accuracy of the ballistic coefficient estimation respectively. Compared with the impact point prediction method based on UF or Extended Kalman Filter （EKF）,the proposed method based on IMM-UF has stronger adaptability and can predict the impact point more quickly and more accurately. The simulation results prove the validity of the method.