Minutes for UA MOPS meeting, March 30, 2006
In attendance:
Pasquale Tricarico, Kobus Barnard, Alon Efrat, Bongki Moon, Scott Morris,
Praveen Rao
Main points:
1. We were joined by Pasquale Tricarico. He is an Associate Research Scientist
at the Planetary Science Institute here in Tucson, AZ. He is the main
developer of the open source orbit fitting and visualization code ORSA.
2. Kobus explained our overall scheme for matching detections and fitting
orbits.
3. Pasquale provided mathematical details of how he approaches the
general orbit/detection association problem. His methods are similar to the
tracklet approach of Jeremy Kubica et al. He relies heavily on having an
accurate velocity vector with each detection.
4. Pasquale described some of the goals he has for the [EMAIL PROTECTED]
project. It seems it might fit well with LSST's outreach requirements.
5. The UA group (Scott) would prefer to use Pasquale's ORSA code (C++)
for fitting orbits since it is more reliable and easier to understand.
However, tests have shown that it is prohibitively slow. Pasquale commented
that he should be able to speed it up to compete with OrbFit, but that it is
not currently a priority.
6. Praveen presented some preliminary results comparing greedy graph
matching to the Hungarian method. There are issues with the threshold and
phantom settings that need to be worked out before a proper comparison is
possible.
7. Pasquale agreed to come again next week to continue our discussion. The
beginning of next week's meeting will also be a telecom with CMU/U.Pitt, but
there should be room for both activities.
--
Kobus Barnard
[EMAIL PROTECTED]
http://kobus.ca
730 Gould Simpson, U. Arizona (520-621-4237)
_______________________________________________
LSST-data mailing list
[email protected]
http://www.lsstmail.org/mailman/listinfo/lsst-data
