正态–伽马非线性雷达扩展目标跟踪滤波器

Normal-gamma nonlinear radar extended target tracking filter

中文关键词  扩展目标  异常噪声  正态–伽马分布  增广非线性量测

英文关键词  extended target  abnormal noise  normal-gamma distribution  augmented nonlinear measurement

中文摘要

      针对复杂异常噪声下的非线性扩展目标跟踪问题,本文提出了一种正态–伽马非线性雷达扩展目标跟踪滤 波器. 首先,对传统扩展目标建模方式进行了优化,将运动状态与扩展状态显式且独立的进行表达;其次,在量测噪 声为异常噪声的情形下,引入辅助变量表示量测噪声协方差的不确定性,并将其建模为伽马分布;同时,考虑到雷达 量测的特殊性,对原有的非线性量测进行了增广,以减少非线性量测在转变过程中的信息丢失问题;最后,构造相应 的扩展目标与群目标跟踪仿真实验,并通过高斯–瓦瑟斯坦距离对扩展目标跟踪多特征联合估计性能进行评估,验 证了本文所提算法的有效性与稳定性.

英文摘要

      For the nonlinear extended target tracking with complex anomalous noise, a normal-gamma nonlinear radar extended target tracking filter is proposed in this paper. Firstly, the traditional extended object modeling method is opti mized, and the motion state and the extended state are formulated explicitly and independently. Furthermore, when the measurement noise is a abnormal noise, auxiliary variables are introduced to represent the uncertainty of the measurement noise covariance, which is a gamma distribution. Additionally, considering the particularity of radar measurement, the original nonlinear measurement is augmented to reduce the problem of information loss during the measurement transfor mation. Finally, simulation experiments of extended target and group target tracking are conducted, and the multi-feature joint estimation performance of extended target tracking is assessed using the Gauss-Wasserstein distance, which verifies the effectiveness and stability of the proposed algorithm.