i am working on my postgrad thesis, and i am going no where in it. i am working on kalman filter which is very difficult for me to design. i need to know what i should do if i have few almost 7 differential equations and i have to design a multiple model kalman estimator.
Hi Noman,
First, you write your models in state-space form, then design Kalman filter for each model individually. You need to decide whether you need a nonlinear Kalman filter or a linear Kalman filter is sufficient for your application. Designing Kalman filter from state-space models is pretty straight-forward.
I learned Kalman filtering from the following book: "Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches" by Dan Simon.
After you are done with designing Kalman filters for your multiple models, you can think of a switching logic between filters. You may think of Fuzzy Logic for the switching between your multiple filters.
Thank you so much sir Esmaeil Jahanshahi . i need further help in designing state space model of suspension system because it is quite complex and having almost 7 differential equations.
i have modeled my system into state space model and now i have to work with kalman filter multiple model. any suggestion?
- Badges
- Science topic
- Similar topics
- Mathematics
- Applied Mathematics
- Differential Equations