Pareto Optimal Filter Design for Nonlinear Stochastic Fuzzy Systems via Multiobjective H /H Optimization

This paper is concerned with the multiobjective H 2 /H ∞ filtering design problem in nonlinear signal processing, which can be approximated by a Takagi-Sugerno (T-S) fuzzy signal system. We propose a multiobjective filter design to estimate state variables from noisy measurements for nonlinear signa...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on fuzzy systems Vol. 23; no. 2; pp. 387 - 399
Main Authors: Chen, Bor-Sen, Lee, Hsin-Chun, Wu, Chien-Feng
Format: Journal Article
Language:English
Published: IEEE 01.04.2015
Subjects:
Filtering
Filtering algorithms
Noise
Noise measurement
Pareto optimization
Stochastic processes
multiobjective problem (MOP)
multiobjective fuzzy filter
Linear matrix inequality (LMI)
pareto optimal solutions
Takagi–Sugerno (T–S) fuzzy model
ISSN:1063-6706, 1941-0034
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper is concerned with the multiobjective H 2 /H ∞ filtering design problem in nonlinear signal processing, which can be approximated by a Takagi-Sugerno (T-S) fuzzy signal system. We propose a multiobjective filter design to estimate state variables from noisy measurements for nonlinear signal systems, and we focus our effort on achieving optimal concurrent performance for H 2 and H ∞ filtering. In general, it is difficult to solve the multiobjective (MO) H 2 /H ∞ fuzzy filter problem directly, and we therefore propose an indirect approach to minimize the upper bounds and transform the MO H 2 /H ∞ filtering problem in to a linear matrix inequality (LMI)-constrained multiobjective problem (MOP). In addition, we propose an LMI-based multiobjective evolution algorithm to efficiently find Pareto optimal solutions for the MOP of multiobjective fuzzy filter design for nonlinear stochastic signal processing. Furthermore, for comparison, we also suggest the MO H 2 /H ∞ filter design problem based on the weighted sum method. Our proposed indirect method can be widely employed to practically address the MO filter design problem in nonlinear signal processing. Finally, a trajectory estimation of reentry vehicle by radar is provided to illustrate the design procedure of the Pareto MO optimal filter.
ISSN:1063-6706
1941-0034
DOI:10.1109/TFUZZ.2014.2312985