According to the official MATLAB documentation, the "linear kalman filter" is used for estimating the state of an object if the estimation system is linear and Gaussian.
You can refer to the MATLAB documentation link below for more information on "linear kalman filter":
For the non-linear systems you can either use "extended kalman filter" or "unscented kalman filter" depending on the use case.
You can refer to the MATLAB documentations below for more information:
- Extended Kalman Filter: https://www.mathworks.com/help/releases/R2024b/fusion/ug/extended-kalman-filters.html#:~:text=If%20your%20system%20is%20nonlinear%2C%20you%20should%20use%20a%20nonlinear%20filter%2C%20such%20as%20the%20extended%20Kalman%20filter%20or%20the%20unscented%20Kalman%20filter%20(trackingUKF).
- Unscented Kalman Filters: https://www.mathworks.com/help/releases/R2024b/fusion/ref/trackingukf.html
- Introduction to estimation filters:https://www.mathworks.com/help/releases/R2024b/fusion/ug/introduction-to-estimation-filters.html
There is a MATLAB example demonstrating how to perform nonlinear state estimation using exteneded kalman function in Simulink for a system which can be accessed using the below link:
Here, instead of the state space matrices(A B C D), the "state transition function" and "measurement model function" can be used.
I hope this helps you.
