MATLAB: Does editing one Kalman filter object change multiple

matlab coder

I create a structure, where one element is a Kalman filter object, then use repmat to create a larger structure containing multiple duplicates of the original structure. I then go through and individually change each index of the structure, but when I change the Kalman filter object state for one index, it is changing it for every index of the structure.

Best Answer

All indices seem to be pointing to the same location in memory for the Kalman filter object. Thus, updating one index of the structure changes the filter object that all indices are pointing to. This results in all the values changing. Separate Kalman filters need to be created for each index, so they will not overwrite each other.

Related Solutions

MATLAB: It is not possible to set a struct with a char field/non numerical field in it as a tunable parameter (Model Parameter Configuration dialog)

When choosing workspace variables to be tunable through the GUI, if the variable is a structure, then the entire structure must be numeric for it to be listed. The doc states that the source list includes:

MATLAB® workspace — Lists all variables in the MATLAB workspace that have numeric values.

It also suggests:

To declare tunable parameters, use Simulink.Parameter objects instead of the Model Parameter Configuration dialog box. See Block Parameter Representation in the Generated Code (Simulink Coder).

The tool looks at each variable in the workspace to check if it is numeric. If the user chooses the variable to be tunable, then the entire variable will be flagged as tunable. In this case, the variable is a structure. This means that the entire structure must be numeric. If just one field is not, then the variable is not numeric and therefore not tunable.

MATLAB: Do I get different results with RESID and FILTER while using System Identification Toolbox 7.1 (R2007b)

This is an expected behavior in System Identification Toolbox 7.1(R2007b). The reason for this behavior is as follows:

m is a dynamic model with one state. So any simulation or prediction calculation requires prescription of initial conditions. When one uses FILTER directly, one is implicitly setting the initial conditions (x0) to zero. Hence with FILTER, the first "error" value is x(1)-x0 = x(1).

However, commands such as PE and RESID estimate initial conditions in such a way that the overall prediction errors are minimized. While RESID does not allow one to change the default handling of initial conditions, PE does. One can enforce zero initial conditions by using:

e2 = pe(m, x', 'init', 'z'); % set initial condition to zero

Then e2 would be same as e. However, if you type:

e3 = pe(m, x')   % default is to estimate ( = 'e') initial conditions

or,

[e3, x0] = pe(m, x', 'init', 'e');

one sees that e3 is same as output of RESID e1. Also, x0 is calculated to be equal to x(1), which leads to zero value for e3(1).

Best Answer

Related Solutions

MATLAB: It is not possible to set a struct with a char field/non numerical field in it as a tunable parameter (Model Parameter Configuration dialog)

MATLAB: Do I get different results with RESID and FILTER while using System Identification Toolbox 7.1 (R2007b)

Related Question