首页> 外文学位 >Sequential Monte Carlo particle filtering for state estimation.
【24h】

Sequential Monte Carlo particle filtering for state estimation.

机译:顺序蒙特卡洛粒子滤波用于状态估计。

获取原文
获取原文并翻译 | 示例

摘要

State estimation is of paramount importance in many fields of Engineering including statistical signal processing, robot localization, and target tracking. Filtering is a way of estimating the state of a system by incorporating noisy observations as they become available online with prior knowledge of the system model. Particle filters are sequential Monte Carlo methods that use a point mass representation of probability densities in order to propagate the required statistical properties for state estimation.; This thesis quantitatively compares the generic and auxiliary particle filtering frameworks using various proposal densities and state characterizations. New particle filtering methods that use the extended and unscented Kalman filters in the auxiliary framework are introduced. All the methods are compared in terms accuracy and robustness.; Synthetic stochastic models that incorporate nonlinear, non-stationary, and non-Gaussian elements are used for the experiments. The newly proposed auxiliary unscented Kalman particle filter is shown to outperform existing nonlinear filters in many of the experiments.
机译:状态估计在工程学的许多领域中至关重要,包括统计信号处理,机器人定位和目标跟踪。过滤是一种通过在系统模型的先验知识就可以在线获取噪声观测值的情况下,通过合并噪声观测值来估计系统状态的方法。粒子过滤器是顺序蒙特卡罗方法,使用概率密度的点质量表示来传播状​​态估计所需的统计特性。本文使用各种提议密度和状态特征定量地比较了通用和辅助粒子过滤框架。引入了在辅助框架中使用扩展和无味卡尔曼滤波器的新粒子滤波方法。比较所有方法的准确性和鲁棒性。实验中使用了包含非线性,非平稳和非高斯元素的合成随机模型。在许多实验中,新提出的辅助无味卡尔曼粒子滤波器均优于现有的非线性滤波器。

著录项

  • 作者

    Smith, Laurence.;

  • 作者单位

    Carleton University (Canada).;

  • 授予单位 Carleton University (Canada).;
  • 学科 Engineering System Science.
  • 学位 M.A.Sc.
  • 年度 2006
  • 页码 150 p.
  • 总页数 150
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 系统科学;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号