基于CKF的多雷达分布式再入弹道目标实时跟踪算法

CKF-based real-time tracking algorithm for multi-radar distributed reentry trajectory target

鉴于分布式数据融合模式具有扩展性强、系统生存能力强的优势,提出一种基于容积卡尔曼滤波(CKF)的多雷达分布式再入弹道目标实时跟踪算法(DCKF)。首先利用统计线性误差传播的方法将CKF算法嵌入扩展信息滤波器得到容积信息滤波器,然后利用一致性算法将多量测值集中式滤波进行分布式等价表示以更新状态估计与误差协方差矩阵,得到DCKF算法。区别于带有数据融合中心的集中式算法,该算法中数据信息的交互仅在具有通信链路的相邻雷达间进行,无数据融合中心。仿真结果表明,DCKF算法能有效提高单雷达测站弹道目标实时跟踪的精度,而且相较于多雷达集中式算法保证了跟踪精度,证明了算法的有效性。

Since the distributed data fusion mode has strong expansibility and powerful system viability,a cubature Kalman filter(CKF)based real-time tracking algorithm for multi-radar distributed reentry ballistic target is proposed.The statistic method for the linear error propagation is used to embed the CKF algorithm into the extended information filter to obtain the cubature information filter,and then the consistency algorithm is used to perform the distributed equivalent representation for the centralized filtering with multiple measured values to update the state estimation and error covariance matrix,so as to obtain the DCKF algorithm.In comparison with centralized algorithm with data fusion center,the data information interaction in the proposed algorithm is carried out between the adjacent radars with communication link,and the algorithm has no data fusion center.The simulation results prove that the DCKF algorithm can improve the single-radar station′s real-time tracking accuracy of ballistic target,and ensure the tracking accuracy similar to multi-radar centralized algorithm,and its effectiveness is verified.