NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Back to Results
Nonlinear consider covariance analysis using a sigma-point filter formulationThe research reported here extends the mathematical formulation of nonlinear, sigma-point estimators to enable consider covariance analysis for dynamical systems. This paper presents a novel sigma-point consider filter algorithm, for consider-parameterized nonlinear estimation, following the unscented Kalman filter (UKF) variation on the sigma-point filter formulation, which requires no partial derivatives of dynamics models or measurement models with respect to the parameter list. It is shown that, consistent with the attributes of sigma-point estimators, a consider-parameterized sigma-point estimator can be developed entirely without requiring the derivation of any partial-derivative matrices related to the dynamical system, the measurements, or the considered parameters, which appears to be an advantage over the formulation of a linear-theory sequential consider estimator. It is also demonstrated that a consider covariance analysis performed with this 'partial-derivative-free' formulation yields equivalent results to the linear-theory consider filter, for purely linear problems.
Document ID
20060043648
Document Type
Conference Paper
External Source(s)
Authors
Lisano, Michael E. (Jet Propulsion Lab., California Inst. of Tech. Pasadena, CA, United States)
Date Acquired
August 23, 2013
Publication Date
February 4, 2006
Subject Category
Mathematical and Computer Sciences (General)
Report/Patent Number
AAS 06-035
Meeting Information
29th Annual AAS Guidance and Control Conference(Breckenridge, CO)
Distribution Limits
Public
Copyright
Other
Keywords
covariance
unscented filters
Inertial measurement unit (IMU) filters
sigma-point formulations