Here estimation of charge is carried out using extended Kalman filtering, I am unclear how we should model the uncertainties involving the battery. How can one incorporate them in our state space ekf model for estimation?
I am unable to get a correct estimation. I am using Matlab scripting to realize the EKF. I am not sure why the script fails to estimate accurately.
Can anyone help me to find out where I am going wrong with reference to the m script attached.
How should we we choose noise parameter Q and R here?
Kindly can someone explain to me what exactly should be considered in these places?
Replies will be much appreciated.
In Bayesian framework of filtering the uncertainties are generally modeled with additive noise. Generally the noise is white Gaussian. You have failed to estimate the states. The reasons may be as follows:
(i) The system is not modeled properly
(ii) EKF is not implemented properly. Instead of any inbuilt code you may write your own code.
(iii) With EKF it is not possible to estimate the states satisfactorily. In that case you may try other advanced filter such as UKF, CKF, CQKF, Particle filter, adaptive grid filter etc.
http://www.tutorialpoint.org/Filter/nonlinear-estimation-literature-review-page1.html
- Badges
- Science topic