ESAC is responsible for developing and running AGIS, the software that computes the global astrometric parameters for the Gaia mission. The design and validation of Gaia global astrometric mission requires to be able to run simulations that include complex calibration issues. The current state of the art is AgisLab. This code is proprietary of DPAC, the scientific consortium processing the Gaia data and responsible for the publication of the final star catalogue.
GaiaLab project is open source, developed by students and going some steps further in order to expose some of the global astrometric issues to a larger community.
For last stable release:
pip install GaiaLab
For the latest code:
git clone https://github.com/bombrun/GaiaLab cd GaiaLab pip install -r requirements.txt
Not yet implemented: For also the notebooks add:
pip install -r notebook_requirements.txt
and enjoy the notebooks as well!
What is implemented
The first version will be based on a very simple model :
- single source
- one ccd
- circular satellite orbit
- Newtonian physic (no relativity)
The second version contain:
- As many source as we like (max ~1000 for rapid use)
- Two telescopes (gaia-like)
- Circular satellite orbit
- Newtonian physics
- Solver with source update (attitude update ongoing)
About the theory
The project will make used of the technical notes written by Lennart Lindegren http://www.astro.lu.se/~lennart/Astrometry/TN.html
In particular the following notes and papers:
- (Lindegren, SAG-LL-14)
- (Lindegren, SAG-LL-30)
- (Lindegren, SAG-LL-35)
- The astrometric core solution for the gaia mission, overview of models, algorithms, and software implementation, L.Lindegren et al.
For contributors and curious:
The codestyle tries to follow PEP8 guidelines, for example using linter 2.2 codestyle package. (see https://atom.io/packages/linter as of 20.09.2018)
The documentation has been created with sphinx. It can be build using
make ***with *** replaced with the kind of documentation you like (e.g.
The online documentation is hosted and linked to the github code through the use of readTheDocs. See https://readthedocs.org/ for more infos.
Link to the presentation of December 14th 2018
link (as of 14/12/2018):