基于SMSS-UKF的机载单站无源定位算法

发布时间：2018-01-04 18:21

本文关键词：基于SMSS-UKF的机载单站无源定位算法　出处：《现代雷达》2017年04期 　论文类型：期刊论文

【摘要】：基于机载单站无源定位中常用的扩展卡尔曼滤波具有近似线性方程的雅可比矩阵计算困难、不敏卡尔曼滤波(UKF)具有计算量较大的问题,提出了基于改进的变尺度最小斜度不敏卡尔曼滤波(SMSS-UKF)的机载单站无源定位算法,该方法采用最小斜度采样策略进行Sigma点采样,减少Sigma点提高计算效率,利用变尺度不敏变换克服了采样点非局部效应问题。同时,针对大部分定位跟踪模型中状态方程为线性方程的特点,依据在线性状态方程情况下的贝叶斯理论,运用卡尔曼滤波状态预测的方法进行UKF的最优状态预测,使状态预测避免了不敏变换的数值近似误差和Sigma采样点计算的复杂性,提高了算法的运行效率和滤波精度。仿真实验的结果证明了SMSS-UKF滤波算法的有效性。
[Abstract]:It is difficult to calculate Jacobian matrix based on extended Kalman filter, which is commonly used in airborne single station passive positioning, and unsensitive Kalman filter (UKF) has the problem of large computational complexity. An airborne single station passive location algorithm based on improved variable scale minimum slope unsensitive Kalman filter (SMSS-UKF) is proposed. The minimum slope sampling strategy is used to sample Sigma points. Reducing the Sigma points to improve the calculation efficiency, using the variable scale insensitive transformation to overcome the sampling point non-local effect problem, at the same time, in order to most of the localization and tracking model, the equation of state is a linear equation. According to the Bayesian theory in the case of linear state equation, the Kalman filter state prediction method is used to predict the optimal state of UKF. The state prediction avoids the numerical approximation error of the insensitive transformation and the complexity of the Sigma sampling point calculation. The operation efficiency and filtering accuracy of the algorithm are improved. The simulation results show that the SMSS-UKF filtering algorithm is effective.
【作者单位】：空军工程大学航空航天工程学院;
【基金】：航空科学基金资助项目(20152096019)
【分类号】：TN95
【正文快照】： 0引言现代空战中,机载单站无源定位系统[1]具有较好的隐蔽性,并且具有设备数量少、作用距离远、机动性能好等优点。无源定位体制包括测向定位[2]、测相位差变化率定位[3-4]、模糊相位差直接定位[5]等。其中,测相位差变化率的定位方法成本低、定位精确可靠,而且避免了相位校准

【相似文献】