一种多扩展目标非线性高斯逆Wishart概率假设密度滤波器

A nonlinear Gaussian inverse Wishart probability hypothesis density filter for multi-extended target tracking

针对非线性多扩展目标跟踪问题,提出了一种基于修正无偏转换量测(modified unbiased converted measurement, MUCM)的高斯逆威沙特概率假设密度(Gaussian inverse Wishart probability hypothesis density, GIW-PHD)滤波器.首先,该方法利用MUCM将雷达非线性量测转换为笛卡尔坐标系下的伪线性量测,并用统计方法得到转换量测误差的协方差矩阵.然后,给出MUCM-GIW-PHD滤波算法的具体实现过程,继而在非线性条件下对多扩展目标的运动参数和形状参数进行联合估计.最后,仿真实验验证了算法的有效性.

Aimed at the problem of nonlinear multi-extended target tracking, a Gaussian inverse Wishart probability hypothesis density (GIW-PHD) filter is proposed in this article based on the modified unbiased converted measurement (MUCM). First, by using MUCM method, the radar’s nonlinear measurement is expressed as pseudo-linear measurement in cartesian coordinate system and a coordinated variance matrix of converted measurement error is obtained by statistical method. Then, the concrete implementation process of MUCM-GIW-PHD filtering algorithm is given to joint estimate the motion parameters and shape parameters of the multi-extended targets under nonlinear condition. Finally, a simulation test has justified the effectiveness of the proposed algorithm.