With other words: Kalman filters enable you exact guessing of the picture where mathematical equations don't do the job because you have too many unknown parameters compared to the number of input sources. I believe the autopilot mailing list archive contains a lot of useful explanations to kalman filtering,
I'm probably wrong, but I got the impression that maybe one person over on the autopilot list (Aaron?) understood Kalman filters well enough to actually impliment them in a real world system. :-)
I like your explanation better than mine though, except I'm not sure about "exact guessing." Maybe I would say "good/best guessing" about the future given noisy and changing inputs, and some knowledge about the error tolerances of the guess. And like you say, the guess is much better than you could generally get with a simple mathematical relationship. Definitely good stuff for real world environments.
Curt. -- Curtis Olson Intelligent Vehicles Lab FlightGear Project Twin Cities [EMAIL PROTECTED] [EMAIL PROTECTED] Minnesota http://www.menet.umn.edu/~curt http://www.flightgear.org
_______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
