satdatagen: a Python Library for Satellite Sensor Task Scheduler Support

Abstract

The number of objects in Earth’s orbit is increasing rapidly, raising urgency for intensified observations of satellites and other resident space objects (RSOs) to manage space traffic and prevent collisions. Current methods for RSO detection and tracking rely on ground-based and space-based observatories with optical or radar sensors, but these telescopes require complex scheduling to achieve surveillance of all objects. Previous works have implemented scheduling algorithms and machine learning models that optimize the assignment of tasks to the sensors for RSO observations. However, prior methodologies rely on different datasets, making it hard to make comparisons across methods. This paper presents satdatagen: a software package that generates datasets that can be used as inputs to sensor task schedulers. The datasets generated from the satdatagen library are intended to be used as a baseline input to satellite sensor task schedulers. The datasets contain information about every satellite that passes in view of the sensor such as its angle of altitude and its brightness. Additionally, actual cloud cover data is included for optical telescopes that need to take visibility into account while scheduling observations. satdatagen is simple to use, and does not require excess outside knowledge from developers of scheduling tools.