基于统计线性回归的粒子滤波方法

郭文艳; 韩崇昭

doi:10.3724/SP.J.1146.2007.01784

基于统计线性回归的粒子滤波方法

doi: 10.3724/SP.J.1146.2007.01784

郭文艳^{①②;韩崇昭},
韩崇昭

基金项目:

国家863计划项目(2006AA01Z126)和国家973计划项目(2007CB 311006)资助课题

计量
- 文章访问数: 2881
- HTML全文浏览量: 70
- PDF下载量: 1398
- 被引次数: 0
出版历程
- 收稿日期: 2007-11-15
- 修回日期: 2008-04-15
- 刊出日期: 2008-08-19

Particle Filter Algorithm Based on Statistical Linear Regression

Guo Wen-yan^{①②;韩崇昭},
Han Chong-zhao

摘要

摘要: 针对非线性、非高斯系统的状态估计问题，该文提出了一种基于统计线性回归的粒子滤波算法。在该算法中，首先对非线性函数基于统计线性回归展开，并利用高斯积分估计回归系数，依此产生重要性密度函数。该密度函数融入了最新的观测信息，扩大了与系统真实后验密度的重叠区域。理论分析和实验结果表明，该算法具有较高的估计精度，与一般的粒子滤波算法相比，有较好的稳定性和较低的计算量。
- 粒子滤波; 状态估计; 统计线性回归; 高斯-厄米特积分; 重要性密度函数
Abstract: In this paper, a new particle filter based on Statistical Linear Regression (SLR) is proposed for the state estimation of non-Gauss nonlinear systems. In the new algorithm, the importance density function of particle filter is generated by linearizing the nonlinear function using statistical linear regression through a set of Gauss-Hermite quadrature points estimating regression coefficient. The density function integrates the new observations into system state transition and extends the overlap fields with true posterior density. The simulation shows that the new algorithm not only has high estimation accuracy but also has better stability and less computation amount than the PF.

HTML全文

参考文献(1)

Shalom Y B, Li X R, and Rubarajan T. Estimation withApplications to Tracking and Navigation [M]. New York:Wiley, 2001: 60-80.[2]Gelb A. Application of Optimal Estimation [M]. Cambridge,MA: MIT Press, 1974: 25-50.[3]Press W H, Teuklsky S A, and Vetterling W T, et al..Numerical Recipes in C 2nd ed [M]. Cambridge, U.K.:Cambridge Univ. Press, 1992: 120-140.[4]Ito K and Xiong K. Gaussian filters for nonlinear filteringproblems [J].IEEE Trans. on Automatic Control.2000,45(5):910-927[5]Cappe O, Godsill S J, and Moulines E. An overview ofexisting methods and recent advances in sequential MonteCarlo [J].Proc. IEEE.2007, 95(4):899-924[6]Arasaranam I, Haykin S, and Elliott R J. Discrete-timenonlinear filtering algorithms using Gauss-Hermitequadrature [J].Proc. IEEE.2007, 95(5):953-974[7]袁则剑, 郑南宁, 贾新春. 高斯-厄米特粒子滤波器[J]. 电子学报, 2003, 31(7): 970-973.Yuan Ze-jian, Zheng Nan-ning, and Jia Xin-chun. The Gauss-Hermite particle filter [J]. Acta Electronica Sinica, 2003,31(7): 970-973.[8]Uhlmann S J and Durrant-Whyte H F. A new method for thenonlinear transformation of means and covariances in filtersand estimatiors[J].IEEE Trans. on Automatic Control.2000,45(3):477-482[9]Doucet A. On sequential simulation based methods forBayesian filtering [ EB/OL ] . available from http ://www.researchindex. com.[10]梁军利, 杨树元, 曲超等. 一种新的基于数值积分的粒子滤波算法[J].电子与信息学报.2007, 29(6):1369-1372浏览[11]Arulampalam M S, Maskell S, and Gordon N. A tutorial onparticle filters for online nonlinear/non-Gaussian Bayesiantracking [J]. IEEE Trans. on Aerospace and ElectronicSystems, 2002, 55(2): 174-188.[12]Merwe R D and Doucet A. The Unscented Particle filter.Advances in neural Information Processing Systems, MIT,2000: 123-165.

施引文献

资源附件(0)

访问统计