舰船电子工程Issue(12)：47-49,61,4.DOI:10.3969/j.issn1672-9730.2014.12.012
粒子滤波和无轨迹粒子滤波算法比较
Comparison of Particle Filter and Unscented Particle Filter Algorithms
摘要
Abstract
To resolve the underwater acoustic position precision and tracking ,this paper establishes a dynamic nonlinear non‐Gaussian model .The particle filter directly employs the state transition prior distribution function which does not include the latest measuring information as an importance density function to approximate the posterior density function .The un‐scented paticle filter brings in resampling technique ,develops the importance density function by unscented transformation that is more close to the posterior density .Simulation results demonstrate that paticle filter and unscented paticle filter all can increase the precision of target position and tracking ,but there is higher estimation precision with the latter .The algorithm of unscented paticle filter is more suitable for engineering pactice .关键词
无轨迹粒子滤波/非线性非高斯/重要性密度函数/重采样/定位跟踪Key words
unscented particle filter/nonlinear non-gaussian/importance density function/resampling/position preci-sion and tracking分类
信息技术与安全科学引用本文复制引用
郑琛瑶,董真杰,张维全..粒子滤波和无轨迹粒子滤波算法比较[J].舰船电子工程,2014,(12):47-49,61,4.基金项目
国家自然科学基金重点项目“基于联合决策与估计的高频超视距雷达信息处理与融合”（编号61135001）资助。 （）
