空中战机的雷达目标信号跟踪精度优化研究

Research on Tracking Accuracy Optimization for Aircraft Radar Signal

机载雷达目标跟踪时,目标存在强非线性运动状态,导致扩展卡尔曼滤波算法性能较差。为解上述问题,将扩展卡尔曼滤波的扩维融合方法与信息滤波相结合,建立扩展信息滤波模型。首先,通过逆协方差形式得到信息状态向量和信息矩阵,将多个低维传感器的观测向量扩展为单个高维观测向量,线性化扩维后的观测矩阵得到雅克比矩阵。然后,求得信息状态向量和信息矩阵的预测值结合融合中心状态信息贡献以及相关的信息矩阵。最后,解出多传感器融合后的状态估计以及状态协方差矩阵。仿真结果表明,与扩展卡尔曼滤加权融合波算法相比,扩展信息滤波(EIF)算法有更低的均方根误差(RMSE)。因此,扩展信息滤波具有很好的跟踪精度,可以为机载雷达目标跟踪优化提供科学依据。

In the process of Aircraft Radar target tracking, the target may in strong nonlinear motion state, which leads to poor tracking performance of Extended Kalman Filtering. In order to resolve the above problem, this paper addressed Extended Information Filtering（ EIF）, which combines the Information Filtering with augmented fusion of Extended Kalman Filtering. Firstly, We obtained information state vector and information matrix by the inverse of co- variance form, extend ts of lower dimension sensors to single higher dimension sensor, and linearized the observe matrix which has been extended. Then, we get multi-sensor fusion information state vector and information matrix after predicted information state vector and predicted information matrix integrated with the information state contribution and associated intbrmation matrix of the fusion center. Finally, we can obtain the state vector and state covariance matrix. The result validates that the extend information filter（EIF） algorithm has better root mean square error（ RMSE）, so it has a good application prospect.