[Bayes++] Estimate robot location from 5 independent estimates each with a varying uncertainty

[Martin N. <mar...@ho...>]

Hello everyone,I am currently doing a team project to build a robot. My rol=
e is to design the decision system. Currently I am working on estimating th=
e robot=92s location. I have inputs from about 5 different systems specifyi=
ng an estimate of the current location. Each of these will contain an (x,y)=
 coordinate and an uncertainty in pixels based upon a common map. The uncer=
tainty for each system varies depending upon where the robot is (e.g. when =
far away from a surveillance camera the uncertainty will be larger than whe=
n close to the camera). Some of these systems can also supply a rotation.I =
need to use a filter to combine these estimates into one with a new uncerta=
inty based upon the input uncertainties.I have read some of the past posts =
on this mailing list and was thinking about using the 'Addative_predict_mod=
el' in Bayes++. Would the noise be the uncertainty I am being given by the =
other systems? I was hoping someone could spend a couple of minutes and jus=
t point me in the right direction.Any help would be greatly appreciated.Tha=
nks,Martin
_________________________________________________________________
Discover the new Windows Vista
http://search.msn.com/results.aspx?q=3Dwindows+vista&mkt=3Den-US&form=3DQBR=
E=