摘要
提出了一种采用遗传算法(GA)优化无味粒子滤波(UPF)的新方法遗传无味粒子滤波器(GAUPF).在无味粒子滤波(UPF)获得比传统粒子滤波(PF)算法更好的重要性采样分布函数的基础上将遗传机制应用于粒子重采样,以进化设计思想克服粒子退化现象,通过优化UPF算法更好地解决了非线性、非高斯领域的目标跟踪问题.仿真结果表明,该算法较好地解决了粒子退化问题,提高了滤波的精确性.
A new method of genetic algorithm to optimize the unscented particles filtering GAUPF has been given. Based on UPF can obtain the better importance sampling distribution function than the traditional PF algorithm, we use the genetic mechanism into particle resample process in order to overcome the particles degradation. GAUPF can solve the problem of object tracking in nonlinear non-gaussian systems be better by optimize UPF. The simulation results show that the proposed methods could solve the problem of particle degeneration and improve the estimated precision.
出处
《天津理工大学学报》
2009年第5期46-49,共4页
Journal of Tianjin University of Technology
基金
天津市高等学校科技发展基金(20071308)
关键词
无味粒子滤波
遗传算法
粒子退化
unscented particles filter （UPF）
genetic algorithm
particle degradation
