摘要
为利用多机实现对目标的快速高精度无源跟踪,提出了一种新的迭代无味卡尔曼滤波(unscented Kalman filter,UKF)算法。所提算法利用随机变量的概率密度函数变换,求得了直接关于目标状态的似然函数,并据此利用最大似然估计迭代求解当前时刻的目标状态,推导了能达到最大似然面的迭代求解准则,将该准则与UKF算法结合得到新的迭代UKF算法。以多机只测角跟踪为例,对所提算法的性能进行仿真分析,仿真结果表明,相对于已有的迭代UKF算法,所提算法具有更好的跟踪性能,实用性强。
A novel iterated unscented Kalman filter （UKF） is proposed to realize fast and high-precision passive tracking using multi-plane. The proposed algorithm gets the likelihood function directly with respect to the target state by the probability density function transformation of the random variable and the state is it- eratively solved using the maximum likelihood estimator based on it. The convergence criterion which can reach the maximum likelihood surface is derived and a novel iterated UKF algorithm is gotten by combining the criterion with the UKF algorithm. The multi plane bearing-only tracking is used as an example to ana lyze the performance of the proposed algorithm. Simulation results indicate that the proposed algorithm has better tracking performance compared with existing iterated UKF algorithms, which indicate its good appli cation prospect.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2014年第2期220-223,共4页
Systems Engineering and Electronics
基金
航空电子系统综合技术重点实验室和航空科学基金(20105584004)联合资助
海军航空工程学院研究生创新基金资助课题
关键词
无源跟踪
迭代无味卡尔曼滤波
最大似然估计
只测角
passive tracking
iterated unscented Kalrnan filter （UKF）
maximum likelihood estimator
bearing-only
