摘要
针对观测方程为非线性,状态方程为线性,且噪声为加性情况下的机动目标跟踪问题,应用RaoBlackwellised UKF(RBUKF)算法滤波并对其进行了防发散处理,得到机动目标跟踪的一种基于防发散RBUKF(AD-RBUKF)算法。对二维机动目标跟踪的仿真结果显示,本算法的跟踪效果明显优于其它两种常用的非线性滤波方法——基本的UKF算法和SPPF算法,仿真结果表明该算法是一种跟踪精度高,且适合工程应用的非线性滤波方法。
Devoted to the problem of maneuvering target tracking under nonlinear observation,linear state and added noise,a kind of algorithm based on anti-divergent RBUKF( AD-RBUKF) of maneuvering target tracking is developed,which uses Rao-Blackwellised UKF( RBUKF) for filter and anti-divergent work.Two-dimensional maneuvering target tracking simulation results show that,the tracking effect of this algorithm is clearly better than the other two commonly used nonlinear filtering method of the basic UKF algorithm and the SPPF algorithm.And the simulation results show that this algorithm is of high tracking accuracy,and it is suitable for nonlinear filtering method for engineering application.
引文
[1]程水英.无味变换与无味卡尔曼滤波[J].计算机工程与应用,2008,44(24):25-35.
[2]刘江,陆明泉,王忠勇.RBUKF算法在GPS实时定位解算中的应用[J].系统工程与电子技术,2009,31(11):2578-2581.
[3]王海霞.舰船组合导航信息综合集成技术应用研究[D].哈尔滨:哈尔滨工程大学,2009.
[4]康叶伟,黄亚楼,孙凤池,等.一种基于RBUKF滤波器的SLAM算法[J].计算机工程,2008,34(1):17-19+29.
[5]张园.卡尔曼滤波及其军事应用[M].北京:国防工业出版社,2016.
[6]敬忠良.神经网络跟踪理论及应用[M].北京:国防工业出版社,1995.