To address the problem of bright commercial satellite streaks in images from the Vera C. Rubin Observatory, we create heuristics for satellite dodging strategies using the survey’s scheduling algorithm. We computationally forecast satellite trajectories to account for the growing satellite population during Rubin operations. Our results help maximize efficiency and quality of this flagship observatory.
Dependencies:
As usual, suggest using anaconda and creating a fresh environement for installing with something like: conda create -n sat-env ; conda activate sat-env
- rubin_sim: Maybe just
conda install -c conda-forge rubin_simFull instructions at: https://github.com/lsst/rubin_sim - shapely:
conda install -c conda-forge shapely - skyfield:
pip install skyfield
Files:
- Compare_sim_runs.ipynb: contains all result plots, including evaluation of dodging efficiency, number of missed exposures, and trade-off between pixel loss and coadded depth.
- test.ipynb & compare_methods.ipynb: contains testing of the satellite streak length method and speed optimization testing
- satellite_util_examples.ipynb: contains plots with satellite simulations
- scheduler_examples.ipynb: contains plots with scheduler simulations
- find_streak_len.py: contains the executable file for calculating streak length and testing dodging efficiency for simulations
- see utils for methods, including calculate streak length methods