Sky Surveys Scheduling Using Reinforcement Learning

Modern cosmic sky surveys (e.g., CMB S4, DES, LSST) collect a complex diversity of astronomical objects. Each of class of objects presents different requirements for observation time and sensitivity. For determining the best sequence of exposures for mapping the sky systematically, conventional scheduling methods do not optimize the use of survey time and resources. Dynamic sky survey scheduling is an NP-hard problem that has been therefore treated primarily with heuristic methods. We present an alternative scheduling method based on reinforcement learning (RL) that aims to optimize the use of telescope resources for scheduling sky surveys.We present an exploration of RL techniques and we implement a Q learning agent through tabular methods. We compare our implementation with standard methods like the Greedy agent and standard frameworks, like Astroplan. We show that tabular-based methods are wholly insufficient for large-scale surveys with large numbers of targets and when long-range planning is required. Future work may be able to use the same environment model and explore approximation methods in order to scale to an input size on the orders of magnitude of an actual sky survey.