NavList:

   

Reply

Geoffrey,


You have not allowed subsequent measurements to move the probability
one way or the other. That is the asumption I referred to. When you
integrate the product of three (or more ) probability densities for a
local region, the shoulders of the distribution become steeper and
narrower than each of the individual distributions. To get the same
probability in a cell away from the center you have to be assuming a
flat density out to the edges.

On Mon, 22 Apr 2002 08:31:45 +0100, Dr. Geoffrey Kolbe wrote:

>
>Rodney, I have made no assumptions regarding the actual error distribution.
>For the purposes of my arguments it can be anything you like.
>
>What I _have_ assumed is that the error in measuring the bearing on a
>particular landmark is equally likely to be to one side of the landmark as
>to the other. Do you have a problem with this?


Rodney Myrvaagnes                                  J36 Gjo/a


"Curse thee, thou quadrant. No longer will I guide my earthly way by thee."  Capt. Ahab

   

Reply