This paper considers the problem of non-Gaussian estimation and dynamic output feedback in both linear and nonlinear settings. Estimation with non-Gaussian process noise, although important in fields such as environmental sampling, is typically problem specific and suboptimal. The approach described here uses the Gaussian mixture model to approximate an unknown non-Gaussian distribution and to employ the Kalman filter and its nonlinear variants: the extended and unscented Kalman filters. The estimation error is bounded analytically and illustrated numerically for linear and nonlinear systems. The estimate is used in an output feedback control to guarantee stability of a closed-loop system.
Non-Gaussian Estimation and Dynamic Output Feedback Using the Gaussian Mixture Kalman Filter
Journal of Guidance, Control, and Dynamics ; 44 , 1 ; 15-24
2021-01-01
Article (Journal)
Electronic Resource
English
Gaussian-Mixture Kalman Filter for Orbit Determination Using Angles-Only Data
Online Contents | 2017
|Gaussian-Mixture Kalman Filter for Orbit Determination Using Angles-Only Data
Online Contents | 2017
|