摘要
传统高斯混合粒子概率假设密度滤波器(Gaussian mixture particle probability hypothesis density filter,GMP-PHDF)采用先验状态转移概率密度作为重要性密度函数,会出现粒子退化问题。而递推更新高斯滤波器依据测量函数梯度渐进式地进行状态更新,可获得更为接近于真实分布的后验估计,但其协方差矩阵易非正定而导致递推中断。对此,本文首先分析平方根递推更新高斯滤波器(square-root recursive update Gaussian filter,SR-RUGF)的实现思路,并给出基于容积卡尔曼滤波(cubature Kalman filter,CKF)的SR-RUGF实现步骤。在此基础上,利用SR-RUGF为GMP-PHDF构建重要性密度函数,进而提出基于平方根递推更新的GMP-PHDF(square-root recursive update GMP-PHDF,SRRU-GMP-PHDF)算法。仿真结果表明,算法可以很好地利用量测信息,获得更高精度的估计结果。
In the traditional Gaussian mixture particle probability hypothesis density filter(GMP-PHDF),the prior state transition probability density is used as a sample important density function,which will lead to a particle degradation problem.The posterior estimation that is more approximate to the real posterior distribution,can be obtained by the incremental state update procedure of the recursive update Gaussian filter according to the gradient of the measurement function,where,however,the non-positive definite covariance matrix will cause recursive interruption.Thus,the implementation idea of the square-root recursive update Gaussian filter(SR-RUGF)is analyzed,and the implementation steps of SR-RUGF based on the cubature Kalman filter(CKF)are given subsequently.On this basis,a sample important density function is constructed by using SR-RUGF,based on which a square-root recursive update GMP-PHDF(SRRU-GMP-PHDF)is derived.Simulation results demonstrate that the proposed algorithm can assimilate the measurement information commendably and obtain estimation results with higher accuracy.
引文
[1]BA H X,CAO L,HE X Y,et al.Modified joint probabilistic data association with classification-aided for multi-target tracking[J].Journal of Systems Engineering and Electronics,2008,19(3):434-439.
[2]REID D B.An algorithm for tracking multiple targets[J].IEEE Trans.on Automatic Control,1979,24(6):843-854.
[3]STREIT R L,LUGINBUHL T E.Maximum likelihood method for probabilistic multi-hypothesis tracking[C]∥Proc.of the SPIE Signal and Data Processing of Small Targets,1994,2235:5-7.
[4]MAHLER R.Multi-target Bayes filtering via first-order multitarget moments[J].IEEE Trans.on Aerospace and Electronic systems,2003,39(4):1152-1178.
[5]VO B N,SINGH S,DOUCET A.Sequential Monte Carlo methods for multi-target filtering with random finite sets[J].IEEE Trans.on Aerospace and Electronic Systems,2005,41(4):1224-1245.
[6]王利伟,司伟建,曲志昱.增强型SMC-PHD多目标跟踪算法[J].系统工程与电子技术,2015,37(10):2205-2211.WANG L W,SI W J,QU Z Y.Improved SMC-PHD algorithm for multiple targets tracking[J].Systems Engineering and Electronics,2015,37(10):2205-2211.
[7]VO B N,MA W K.The Gaussian mixture probability hypothesis density filter[J].IEEE Trans.on Signal Processing,2006,54(11):4091-4104.
[8]KOTECHA J H,DJURIC P M.Gaussian particle filtering[J].IEEE Trans.on Signal Processing,2003,51(10):2592-2601.
[9]CLARK D,VO B T,VO B N.Gaussian particle implementations of probability hypothesis density filters[C]∥Proc.of the IEEE Aerospace Conference,2007:1-11.
[10]MERWE R,DOUCET A,FREITAS N,et al.The unscented particle filter[C]∥Proc.of the 13th International Conference on Neural Information Processing Systems,2000:563-569.
[11]HOU S Y,HUNG H S,KAO T S.Extended Kalman particle filter angle tracking(EKPF-AT)algorithm for tracking multiple targets[C]∥Proc.of the International Conference on System Science and Engineering,2010:216-220.
[12]康莉,谢维信,黄敬雄.基于unscented粒子滤波的红外弱小目标跟踪[J].系统工程与电子技术,2007,29(1):1-4.KANG L,XIE W X,HUANG J X.Tracking of infrared small target based on unscented particle filtering[J].Systems Engineering and Electronics,2007,29(1):1-4.
[13]LU C G,FENG X X,LEI Y,et al.A novel particle filter for nonlinear non-Gaussian estimation[C]∥Proc.of 3rd International Workshop on Intelligent Systems and Applications,2011:1-5.
[14]HUBER M F,HANEBECK U D.Gaussian filtering for polynomial systems based on moment homotopy[C]∥Proc.of 16th IEEE International Conference on Information Fusion,2013:1080-1087.
[15]ZANETI R.Recursive update filtering for nonlinear estimation[J].IEEE Trans.on Automatic Control,2012,57(6):1481-1490.
[16]张勇刚,王刚,黄玉龙,等.递推更新高斯粒子滤波[J].控制理论与应用,2016,33(3):353-360.ZHANG Y G,WANG G,HUANG Y L,et al.Recursive update Gaussian particle filter[J].Control Theory&Application,2016,33(3):353-360.
[17]何友,修建娟,关欣.雷达数据处理及应用[M].3版.北京:电子工业出版社,2013.HE Y,XIU J J,GUAN X.Radar data processing with applications[M].3rd ed.Beijing:Publishing House of Electronics Industry,2013.
[18]VAN DER M R,WAN E A.The square-root unscented Kalman filter for state and parameter-estimation[C]∥Proc.of the IEEE International Conference on Acoustics,Speech,and Signal Processing,2001:3461-3464.
[19]刘华,吴文,王世元.基于平方根CKF的多传感器序贯式融合跟踪算法[J].系统工程与电子技术,2015,37(7):1494-1498.LIU H,WU W,WANG S Y.Multi-sensor sequential fusion tracking algorithm based on square-root cubature Kalman filter[J].Systems Engineering and Electronics,2015,37(7):1494-1498.
[20]张召友,郝燕玲,吴旭.3种确定性采样非线性滤波算法的复杂度分析[J].哈尔滨工业大学学报,2013,45(12):111-115.ZHANG Z Y,HAO Y L,WU X.Complexity analysis of three deterministic sampling nonlinear filtering algorithms[J].Journal of Harbin Institute of Technology,2013,45(12):111-115.
[21]VO B N,VO B T,SCHUHMACHER D.A consistent metric for performance evaluation of multi-object filters[J].IEEE Trans.on Signal Processing,2008,56(8):3447-3457.
