ADAPTIVE SPLIT-MERGE ALGORITHM FOR GAUSSIAN MIXTURE MODELS TO SOLVE THE KOLMOGOROV EQUATION

被引：0

作者：

Vishwajeet, Kumar ^{[1
]}

Singla, Puneet ^{[1
]}

机构：

[1] Univ Buffalo, Dept Mech & Aerosp Engn, Buffalo, NY 14260 USA

来源：

ASTRODYNAMICS 2013, PTS I-III | 2014年 / 150卷

关键词：

POLYNOMIAL-CHAOS; STATE;

D O I：

暂无

中图分类号：

V [航空、航天];

学科分类号：

08 ; 0825 ;

摘要：

Number of components in a Gaussian mixture model plays an important role in its accuracy and computational complexity. New adaptive split-merge technique is introduced in this paper based on the minimization of error in the solution of Fokker Planck Kolmogorov Equation. We also discuss the effect of splitting/merging of few components on the weights of other components. A single Gaussian component at initial time is split over time to account for the change in the probability density function of the states of the system.

引用

页码：977 / 988

页数：12

共 20 条

[1] Abramov RV, 2010, COMMUN MATH SCI, V8, P377
[2] NONLINEAR BAYESIAN ESTIMATION USING GAUSSIAN SUM APPROXIMATIONS
ALSPACH, DL
SORENSON, HW
[J]. IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1972, AC17 (04) : 439 - &
[3] [Anonymous], 2007, Fundamentals of Astrodynamics and Applications
[4] [Anonymous], 1989, Methods of solution and applications
[5] Clark E., 1999, AC SPEECH SIGN PROC, V6
[6] Progressive bayes: A new framework for nonlinear state estimation
Hanebeck, UD
Briechle, K
Rauh, A
[J]. MULTISENSOR, MULTISOURCE INFORMATION FUSION: ARCHITECTURES, ALGORITHMS, AND APPLICATIONS 2003, 2003, 5099 : 256 - 267
[7] Jazwinski A.H., 1979, STOCHASTIC PROCESSES
[8] Julier SJ, 2002, P AMER CONTR CONF, V1-6, P887, DOI 10.1109/ACC.2002.1023128
[9] Unscented filtering and nonlinear estimation
Julier, SJ
Uhlmann, JK
[J]. PROCEEDINGS OF THE IEEE, 2004, 92 (03) : 401 - 422
[10] Kalman R.E., 1960, NEW APPROACH LINEAR, DOI [DOI 10.1115/1.3662552, 10.1115/1.3662552]

← 1 2 →