Hello all,
I've posted a new development version at
http://www.projectpluto.com/pluto/devel/find_orb.htm
(note that the site for the development version has moved).
I recommend upgrading to this version. It includes some bug
fixes, and a considerably better initial orbit determination (IOD)
routine. As has been mentioned on this list, Find_Orb had a bad
habit of picking out very unlikely orbits for some short arcs.
In these cases, it would pick an orbit that fit the observations
well, with small residuals; but it would pick an "exotic" orbit
(high-inclination, high-eccentricity, etc.) when some mundane,
main-belt orbits fit about as well.
Find_Orb now looks a little bit deeper and considers common
orbits first, before looking for exotic ones. There's still some
work to be done here, but it's much better than it was.
Also: to the source code for creating console versions of
Find_Orb in Linux, OS/X, and Windows, I have added source
code for a new 'fo' program. This is a non-interactive flavor
of Find_Orb. You run it, and it simply determines orbital elements
and writes them out to files without requiring human intervention.
As always, bug reports and comments are solicited.
-- Bill