摘要
针对复合K噪声干扰下目标跟踪系统中出现的强非线性非高斯问题,在给出一种复合K噪声统计模型的基础上,提出将容积粒子滤波(CPF)与无迹粒子滤波(UPF)两种算法应用在典型目标跟踪系统中,并对算法的跟踪性能进行了仿真分析。实验结果表明,CPF,UPF两种算法均能有效跟踪复合K噪声下的运动目标;其中,CPF算法表现出更高的跟踪精度和更好的实时性,且具有更低的算法设计复杂度。
Aimed at the strong nonlinear and non-Gaussian problems of target tracking system under compound K noise jamming, the Cubature Particle Filter（ CPF） algorithm and Unscented Particle Filter（ UPF） algorithm are applied in typical target tracking systems based on the compound K noise statistical model, and the tracking performance of the algorithm are analyzed through simulation. The experimental results illustrate that： 1） Both the CPF and UPF algorithms have good performance in tracking moving target under condition with compound K noise; and 2） Compared with UPF, the CPF has higher tracking accuracy,better real-time performance, and lower complexity in the algorithm design.
出处
《电光与控制》
北大核心
2016年第5期1-5,共5页
Electronics Optics & Control
基金
国家自然科学基金(61203007)
关键词
目标跟踪
容积粒子滤波
无迹粒子滤波
非线性非高斯
复合K噪声
target tracking
Cubature Particle Filter（CPF）
Unscented Particle Filter（UPF）
nonlinear non-Gaussian
compound K noise
