Combined Parameter and State Estimation in Particle Filtering

In this paper, an adaptive estimation algorithm is proposed for non-linear dynamic systems with unknown static parameters based on combination of particle filtering. The estimates of static parameters are obtained by state samples and maximum-likelihood estimation in particle filtering, and the stoc...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaojun Yang, Kunlin Shi, Tao Huang, Keyi Xing
Format: Conference Proceeding
Language:English
Subjects:
Adaptive estimation
Adaptive filters
Information filtering
Information filters
Maximum likelihood estimation
Nonlinear dynamical systems
Parameter estimation
particle filtering
Recursive estimation
sequential Monte Carlo
State estimation
Yttrium
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, an adaptive estimation algorithm is proposed for non-linear dynamic systems with unknown static parameters based on combination of particle filtering. The estimates of static parameters are obtained by state samples and maximum-likelihood estimation in particle filtering, and the stochastic approximation is used to approximate the gradient of cost function. The proposed algorithm achieves combined state and parameters estimation. Simulation result demonstrates the efficiency of the algorithm.
ISSN:1948-3449
1948-3457
DOI:10.1109/ICCA.2007.4376514