Discrete Time Kalman Filter for Higher Order Systems

[Calorified]

Hello there,

 

I tried to implement the discrete time Kalman filter in a closed loop estimator scenario based on the "CDEx LQG with Linear Simulation" example located in 

 

             `C:\Program Files (x86)\National Instruments\LabVIEW 2015\examples\Control and Simulation\Control Design\State-Space Synthesis`

 

that ships with LabVIEW. Only difference is my application is a 4 x 4 state matrix.

 

The thing is the kalman filter output, \( y(t)(k|k) \) and output estimate, \( \hat{y}(k|k)\) do not get updated.

My vi.png is here:

 

 

I know it says somewhere on the help pages that the KF vi needs a second-order process components. But does this mean the vi would not work with higher order systems?

 

Would appreciate your input.

Edited September 28, 2015 by Calorified

[Calorified]

Turns out my intuition was right. The discrete KF vi will not accommodate higher-order stochastic state space systems. I confirmed this through the LaVIEW PSC engineers yesterday.

 

If anyone is thinking of a big state space structure as I was, the easy way out might be to build your own KF application.