arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1303.1205math.NAcs.NA
收藏

多变量反馈粒子滤波器

Multivariable Feedback Particle Filter

University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);University of Florida at Gainesville(佛罗里达大学盖恩斯维尔分校)

Tao Yang, Richard S. Laugesen, Prashant G. Mehta, Sean P. Meyn

AI总结:

本文通过创新结构避免重要性采样,提出多变量反馈粒子滤波器,改进增益计算,并建立一致性、良态性及有限元算法。

摘要:

在最近的工作中,通过结合传统非线性滤波和平均场博弈形式主义的创新结构,表明可以在粒子滤波中避免重要性采样。由此产生的反馈粒子滤波器(FPF)提供了显著的方差改进;特别是,该算法可以应用于不稳定的系统。该滤波器需要预先计算滤波增益。本文描述了在一般多变量设置中计算增益的新表示和算法。主要贡献是:(i) 改进了FPF的理论:在多变量设置中建立了一致性,以及相关PDE的良态性以获得滤波增益。(ii) 增益可以表示为一个函数的梯度,该函数正是相关MCMC扩散(Smoluchowski方程)的泊松方程的解。这为MCMC以及近似最优滤波方法(如TD学习)提供了桥梁,这些方法反过来可用于近似增益。(iii) 受泊松方程弱形式的启发,提出了一种Galerkin有限元算法来近似增益。其性能在数值实验中得到了说明。

英文摘要:

In recent work it is shown that importance sampling can be avoided in the particle filter through an innovation structure inspired by traditional nonlinear filtering combined with Mean-Field Game formalisms. The resulting feedback particle filter (FPF) offers significant variance improvements; in particular, the algorithm can be applied to systems that are not stable. The filter comes with an up-front computational cost to obtain the filter gain. This paper describes new representations and algorithms to compute the gain in the general multivariable setting. The main contributions are, (i) Theory surrounding the FPF is improved: Consistency is established in the multivariate setting, as well as well-posedness of the associated PDE to obtain the filter gain. (ii) The gain can be expressed as the gradient of a function, which is precisely the solution to Poisson's equation for a related MCMC diffusion (the Smoluchowski equation). This provides a bridge to MCMC as well as to approximate optimal filtering approaches such as TD-learning, which can in turn be used to approximate the gain. (iii) Motivated by a weak formulation of Poisson's equation, a Galerkin finite-element algorithm is proposed for approximation of the gain. Its performance is illustrated in numerical experiments.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑