Updated Find_Orb posted

Bill Gray Nov 4, 2017

Hello all,

I've posted an update with a variety of changes that may,
I think, be of interest to some of you. As usual, you can
download the current software from

https://www.projectpluto.com/find_orb.htm

If you're among the few building Find_Orb from source code
on Linux, *BSD, or OS/X, directions are at

https://www.projectpluto.com/find_sou.htm

Or you can just use the on-line orbit solver.

The most significant changes in this new version are
described more fully at

https://projectpluto.com/findhist.htm

There are two changes that will probably get the most
attention. If you add the line

COM Combine all

at the top of a file of observations, they'll all be treated
as if they were of a single object, and Find_Orb will attempt
to compute an orbit satisfying all observations. This is quite
useful if, for example, you're trying to decide if two objects
really can be linked. Previously, you had to mess around a bit
with changing the designations so they'd match.

The second fix is a considerable improvement in initial orbit
determination for longer arcs, including those where you have
(for instance) a short arc at one opposition, a gap of years or
decades, and then another short arc, or even just a pair of
observations. Following some comments from Rob Matson about
attempting to link 2013 SM19 to 2000 TF2,

https://groups.yahoo.com/neo/groups/MPML/conversations/messages/33292

I took another try at this issue and came up with a somewhat
more elegant scheme for computing such linkages. Note that this
applies to Windows Find_Orb, the on-line Find_Orb, and Find_Orb
built from source code on Linux or *BSD or OS/X.

There are also some more minor bug fixes and improvements. I've
been doing a fair bit of work on the ability to do precovery searches
(give Find_Orb a big list of images, solve for an orbit for a given
object, and it'll tell you which image(s) you should look at to find
that object). Sort of like SkyMorph, except (I hope) extensible
to any survey willing to release a list of what areas they've imaged
when, and (probably the biggest stumbling block) some means to
access the actual images. So the code is ready, but the surveys
aren't quite ready yet.

-- Bill