Hi All,
I am working on a simple dynamic system with some unknown parameters (i.e. moment of inertia etc). The system is controlled by an closed LQR. I read papers about Kalman filter for state measurement and parameter identification (Ljung's method, that seems should work). What is the best algorithm for my application?
Thanks a lot.
Everett
Kalman filtering is usually better for estimating 'states', i.e., variables changing with time. Seems in your case what's unknown is 'parameters', something more or less constants. So you may want to use parameter identification. There are two kinds of methods to estimating a parameter. One is 'on-line' and are closely related to adaptive control. The other is 'off-line', by which you get a lot of data and use the data to obatin an approximation of the parameter. The approximated value of the parameter is then used to calculate the control gains. My guess is that the last description fits your problem.
HTH,
-RT
Have something to add? Share your thoughts — no account required.
Ask the community — no account required