Fractional order extended Kalman filtering method considering Levy noise
The invention discloses a fractional order extended Kalman filtering method considering Levy noise. According to the method, firstly, a state predicted value and a predicted error covariance initial value are given; secondly, the Levy noise is processed, a measurement noise covariance is computed, a...
Saved in:
Main Authors | , , , , , , , |
---|---|
Format | Patent |
Language | English |
Published |
05.08.2015
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a fractional order extended Kalman filtering method considering Levy noise. According to the method, firstly, a state predicted value and a predicted error covariance initial value are given; secondly, the Levy noise is processed, a measurement noise covariance is computed, and an optimal filtering gain is computed through the obtained measurement noise covariance; thirdly, a state estimation value and an estimation error covariance are computed through the optimal filtering gain; and fourthly, the state predicted value is updated through the state estimation value, and the predicted error covariance is updated through the estimation error covariance. The method can solve the problem of state estimation of a fractional order nonlinear discrete system in non-Gaussian noise, and is easy to combine with existing state estimation software. |
---|---|
Bibliography: | Application Number: CN20151250515 |