粒子滤波算法研究进展与展望

The Development and Prospect of Particle Filtering Algorithm

粒子滤波是一种通过非参数化的Monte Carlo模拟方法实现递推贝叶斯估计的算法。本文对粒子滤波的发展和研究现状进行了阐述,详细介绍和分析了粒子滤波的基本原理、存在的几个关键问题及解决方法,总结归纳出11种主要改进粒子滤波器,同时论述了粒子滤波应用领域。最后对未来发展提出了展望。

Particle filtering is an non-parameterized algorithm via sequential Monte Carlo simulation to actualize bayesian estimation.This paper expatiated the development and the research status of particle filtering at present.Then,introduces and analyses the principle of particle filtering、the existed key problems and countermeasures in detail.Furthermore,it sum up eleven improved methods of particle filtering algorithm.Meanwhile,applications are addressed in this paper.Finally,the prospect of particle filtering is presented.