How to avoid normalization of particle flow for nonlinear filters, Bayesian decisions and transport

被引：3

作者：

Daum, Fred ^{[1
]}

Huang, Jim ^{[1
]}

机构：

[1] Raytheon Co, Woburn, MA 01801 USA

来源：

SIGNAL AND DATA PROCESSING OF SMALL TARGETS 2014 | 2014年 / 9092卷

关键词：

particle filter; nonlinear filter; particle flow; transport problem; extended Kalman filter; Monge-Kantorovich optimal transport; MONTE-CARLO;

D O I：

10.1117/12.2044122

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

We describe four distinct ways to avoid normalization of the probability density for particle flow. We have roughly 20 algorithms to compute particle flow, and the three best algorithms avoid computing the normalization of the conditional probability density of the state. We explain why explicit normalization often spoils the flow. This phenomenon has been noticed by other researchers for completely different applications (e. g., weather prediction), but apparently the benefits of avoiding normalization are not well known.

引用

页数：9