天基预警雷达自适应转换测量卡尔曼滤波

An Adaptive Converted Measurement Kalman Filtering Algorithm for Space-based Early Warning Radar

本文针对天基预警雷达体制下的目标跟踪问题,首先建立了天基预警雷达几何运动模型和跟踪系统模型,然后引入求容积法和噪声协方差调整系数,提出了一种噪声协方差自适应的无偏转换测量卡尔曼滤波算法。该算法采用无偏转换和坐标转换将观测量从东北天极坐标系转换到地心固连直角坐标系,利用求容积法求解地心固连直角坐标系下的量测协方差,利用新息序列调整协方差矩阵Q和R,解决了天基预警雷达目标跟踪中的坐标转换误差和噪声协方差不匹配引起的滤波发散问题,最后通过蒙特卡罗仿真实验,与现有算法进行对比分析,证明了所提算法的有效性。

In order to solve the target tracking problem in the space-based early warning radar system condition, first, this paper establishes a geometrical motion model and a tracking system model of space-based early warning radar; secondly, this paper introduces the idea of cubature and an adjustment coefficient for the covariance matrix, and finally, an adaptive converted measurement Kalman filtering （CMKF） algorithm is put forward. The proposed algorithm transforms the meas- ured polar coordinate data of ENZ coordinate system into ECEF coordinate system by means of Unbiased transformation, and transforms the measurement covariance matrix into ECEF coordinate system by means of cubature, and then use the new in- formation matrix to adjust the covariance matrices Q and R. The proposed algorithm solved the problems of coordinate con- version error and the eovariance matrix mismatching in space-based early warning radar. Finally, compared with the exist- ing algorithms, the simulation results of Monte-Carlo experiments proved that the proposed algorithm improves the problem of filtering divergence and reduces the filtering error, and the new algorithm improves the space-based early warning radar tracking performance.