par Arenou, Frédéric;Babusiaux, Carine 
;Chéreau, Fabien;Mignot, Shan
Référence ESA SP, 576, page (335-342)
Publication Publié, 2005
;Chéreau, Fabien;Mignot, ShanRéférence ESA SP, 576, page (335-342)
Publication Publié, 2005
Article révisé par les pairs
| Résumé : | Because Gaia will perform a continuous all-sky survey at a medium (Spectro) or very high (Astro) angular resolution, the on-board processing needs to cope with a high variety of objects and densities which calls for generic and adaptive algorithms at the detection level, but not only. Consequently, the Pyxis scientific algorithms developed for the on-board data handling cover a large range of application: detection and confirmation of astronomical objects, background sky estimation, classification of detected objects, Near-Earth Objects onboard detection, and window selection and positioning. Very dense fields, where the real-time computing requirements should remain within fixed bounds, are particularly challenging. Another constraint stems from the limited telemetry bandwidth and an additional compromise has to be found between scientific requirements and constraints in terms of the mass, volume and power budgets of the satellite. The rationale for the on-board data handling procedure is described here, together with the developed algorithms, the main issues and the expected scientific performances in the Astro and Spectro instruments. |
