摘要
粒子滤波算法广泛应用于现代的跟踪与定位,它适用于非线性非高斯系统,其算法的性能很大程度上取决于重要性分布的选择。分析粒子滤波的原理,论述蒙特卡罗方法在贝叶斯分析中的应用,同时将蒙特卡罗方法引入粒子滤波算法的实现中,通过仿真实验结果比较分析了改进的粒子滤波算法的性能。
Particle filter is widely used in the tracking and location application,especially it can be used in nonlinear and non-Gaussian system.The performance of particle filter algorithm almost lies on the selection of the proposal importance distribution.This paper analyze the principle of particle filter,and explain the application of MCMC in the Bayes analysis,and using MCMC method to improve the particle filter.The performance of the improved particle filter is given out by simulation results.
出处
《杭州电子科技大学学报(自然科学版)》
2007年第6期52-55,共4页
Journal of Hangzhou Dianzi University:Natural Sciences