基于ML背景参数估计的CDKF-CPHD多目标跟踪算法

发布时间：2018-02-25 21:08

本文关键词： 重尾分布中心差分法幅值信息极大似然估计虚警非线性系统 OSPA距离信杂比　出处：《北京航空航天大学学报》2017年03期 　论文类型：期刊论文

【摘要】：针对低信杂比环境下的多机动目标跟踪问题,提出了一种基于极大似然(ML)背景参数估计的中心差分卡尔曼-势概率假设密度滤波(BE-CDKF-CPHD)算法。算法采用ML法实时估计重尾分布模型参数,计算检测概率和虚警概率。运用极大似然-恒虚警(MLCFAR)算法对信号进行处理,提取有效量测值,将幅值似然函数与势概率假设密度滤波器(CPHD)中的目标位置似然函数相结合,通过中心差分法递归更新得到后验均值与协方差,达到对多机动目标进行跟踪的目的。仿真结果表明,在低信杂比环境中,所提算法提高了跟踪精度与目标数目估计准确度。
[Abstract]:According to the low signal to clutter ratio tracking problem of multiple maneuvering target environment, propose a method based on maximum likelihood (ML) estimation of background parameters of the central difference Calman Cardinalized probability hypothesis density filter (BE-CDKF-CPHD) algorithm. The algorithm uses the ML method to estimate the heavy tailed distribution of model parameters, calculate the detection probability and false alarm probability. The use of maximum likelihood - CFAR (MLCFAR) algorithm for signal processing, extraction of effective measurements, the amplitude of the likelihood function and potential probability hypothesis density filter (CPHD) target position in the combination of the likelihood function, the central difference method recursive update posterior mean and covariance, achieve the goal of tracking for multiple maneuvering targets. The simulation results show that in the low SNR environment, the proposed algorithm improves the tracking precision and the target number estimation accuracy.

【作者单位】：西北工业大学自动化学院;
【基金】：航空科学基金(20152853029)~~
【分类号】：TN713

【参考文献】