I sounds like you really only need a state estimator.  A kalman filter
handles things like variable uncertainty (which is essential for
things like GPS).  It takes quite a lot of computing power (you have
to invert a matrix, with is N^3) and is complicated to get right (some
corner cases can be unstable).

If your system error is constant, then a state estimator is all you
need.  You just need a model of the system (probably just a second
order physics model), and an estimate of the noise (probably hand
tuned).  Any controls book on state space controls should cover state
estimators.

-Dan

On Wed, Aug 17, 2011 at 9:34 PM, Dave <[email protected]> wrote:
...

------------------------------------------------------------------------------
Get a FREE DOWNLOAD! and learn more about uberSVN rich system, 
user administration capabilities and model configuration. Take 
the hassle out of deploying and managing Subversion and the 
tools developers use with it. http://p.sf.net/sfu/wandisco-d2d-2
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to