Does anyone know of a COTS star pattern identification software?

Does anyone know of a COTS star pattern identification software?
Obviously, 'lost in space' will also work (might be more expensive,
though).


References to literature/descriptions of relevant algorithms would also

be very useful.


Note: I am looking for a little more detail than "make triangles out of

star centroids in the image, measure the distance between centroids in
the triangle, then compare with a database of same, generated from a
star catalog (correcting for proper motion)."


Thanks!

Yegor wrote:
Does anyone know of a COTS star pattern identification software?
Obviously, 'lost in space' will also work (might be more expensive,
though).


References to literature/descriptions of relevant algorithms would also

be very useful.


Note: I am looking for a little more detail than "make triangles out of

star centroids in the image, measure the distance between centroids in
the triangle, then compare with a database of same, generated from a
star catalog (correcting for proper motion)."


Ball will apparently gleefully sell you a system, but when I talked
to one of their engineers some years ago he didn't want to be any
more specific than that - evidently they figured that their
details were proprietary. Hint from debriefing the Astro project'
scientists - be really sure of your star catalog's lineage and
the camera's wavelength span of sensitivity if you want to include
an intensity cut for star IDs.

Bill Keel

http://spiff.rit.edu/match/

Michael Richmond

William C. Keel wrote:
Yegor wrote:
Does anyone know of a COTS star pattern identification software?
Obviously, 'lost in space' will also work (might be more expensive,
though).


References to literature/descriptions of relevant algorithms would also
snip
Ball will apparently gleefully sell you a system, but when I talked
to one of their engineers some years ago he didn't want to be any
more specific than that - evidently they figured that their
details were proprietary. Hint from debriefing the Astro project'
scientists - be really sure of your star catalog's lineage and
the camera's wavelength span of sensitivity if you want to include
an intensity cut for star IDs.

Not to mention actually simulating the star tracker in all orientations
to ensure that the catalog is actually dense enough to not have
embarrasing gaps.

William C. Keel wrote:
Ball will apparently gleefully sell you a system, but when I talked
to one of their engineers some years ago he didn't want to be any
more specific than that

I'm surprised; the data sheet for their standard star tracker is at
http://www.ballaerospace.com/pdf/ct601.pdf

The data sheet is pretty old; Ball may have other devices available.
Lockheed-Martin made the star trackers for Spitzer, which are similar
in principle to the Ball CT-601 but deliver sub-arcsecond accuracy.
I'm not sure who made the Chandra star trackers, but it must be
possible to find out.

Of course when dealing with any large aerospace company, bring money.
I don't know the price of a CT-601, but my guess would be 7 figures
and the first digit 1.

Yegor wrote:
Does anyone know of a COTS star pattern identification software?
Obviously, 'lost in space' will also work (might be more expensive,
though).

You might want to ask Henry Spencer ( henry at zoo.utoronto.ca ) what
they used on MOST http://www.astro.ubc.ca/MOST/

The amsat http://www.amsat.org/amsat-new/index.php folks might also
have some recommendations.

"Of course when dealing with any large aerospace company, bring money.
I don't know the price of a CT-601, but my guess would be 7 figures
and the first digit 1."

Yes, that's exactly why I'm posting here instead of just calling
salesmen.