Re: Kalman tuning
From: John Lansberry (john.lansberry_at_jhuapl.edu)
Date: 08/17/04
- Next message: nastaran hendijani: "aprobem in fft computation"
- Previous message: Magnus Sundberg: "Re: how to simplify a summation of exponentials symbolicly/analytically?"
- In reply to: ta: "Kalman tuning"
- Messages sorted by: [ date ] [ thread ]
Date: Tue, 17 Aug 2004 07:52:00 -0400
"ta" <motorpg@yahoo.com> wrote in message
news:d69d0bf.0408131401.40e567bf@posting.google.com...
> Hi,
>
> Other than the measurements, I do not
> have any other truth data. So how do I tell if the tuning is resulting
> in a better filter or not.
Of course, "truth data" rarely exists in the real world. "Truth data"
typically is found only in simulations of the real world. If you know
enough about the underlying dynamics of the system to develop a plant model
for the Kalman filter, then perhaps you can develop a "truth model" in a
simulation, and test your Kalman filter state estimates to "simulation
truth."
John
- Next message: nastaran hendijani: "aprobem in fft computation"
- Previous message: Magnus Sundberg: "Re: how to simplify a summation of exponentials symbolicly/analytically?"
- In reply to: ta: "Kalman tuning"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
