Hi folks,
I've posted a pretty big update to the "development" flavor of Find_Orb, at
http://www.projectpluto.com/pluto/devel/find_orb.htm
A lot has been added. Find_Orb now supports:
-- A new observation filtering method that drops the weight given to
observations with doubtful residuals;
-- A method of ensuring that observatories that over-observe an object,
getting dozens of observations of one object in one night, don't dominate the
solution too much;
-- A check-box to have residuals given to an extra digit;
-- Use of the AstDyS/NEODyS biases and weights, when using the .rwo files
of observations provided on those sites;
-- An alternative orbital element display that lets me pack in a little more
data. Not much more yet, but it's paving the way to display sigmas on-screen.
-- Ability to turn on/off weighting of observations with a check-box.
Details on these are given at the above URL. Please let me know if you've
thoughts on any of this. In particular, the first two items are essentially
experimental. I'm confident that they are better than the current "standard"
practices for handling both problems, but not confident that they are as good
as they could be.
-- Bill