NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: John Karl
Date: 2013 Mar 15, 16:03 -0700
Marcel,
Yup, you have it exactly right.
Your case No. 1 is an estimation problem using sights from an unknown position, but with known LOP probability distributions gathered through many, many, previous measurements, which is your case No. 2.
This is exactly the (apparently) subtle distinction some of us have been trying to point out.
Andrew Nikitin appears to be performing your No. 2 above, but now is tabulating MMPs (symmedians) all determined from the same known location (instead of LOPs) -- it's not an estimation problem, like your case No. 1. Rather it's your case No. 2, determining measurement distributions from known facts. His dot figure looks quite believable with the dot density maximum at the true position. But I don't believe his resulting graph of the probability density -- it should obviously be maximum at the true position, as we can see from the dot pattern.
He, and others Like Nanno and Gary, apparently are plotting some distance related quantity instead of simply the number of events per unit area, versus distance.
JK
----------------------------------------------------------------
NavList message boards and member settings: www.fer3.com/NavList
Members may optionally receive posts by email.
To cancel email delivery, send a message to NoMail[at]fer3.com
----------------------------------------------------------------