Observation Strategy Optimization for Distributed Telescope Arrays with Deep Reinforcement Learning

Abstract Time-domain astronomy is an active research area now, which requires frequent observations of the whole sky to capture celestial objects with temporal variations. In optical band, several telescopes in different locations could form a distributed telescope array images continuously. However, there are millions observe each night, and only limited be used for observation. Besides, observation capacity these would affected by effects, such as background or seeing condition. It necessary develop algorithm optimize strategy arrays according scientific requirements. this paper, we propose novel framework that includes digital simulation environment deep reinforcement learning arrays. Our obtain effective strategies given predefined requirements information. To test performance our algorithm, simulate scenario uses space debris. Results show better results both discovery tracking The proposed paper optimization arrays, Sitian project TIDO project.