Revising the Observation Satellite Scheduling Problem Based on Deep Reinforcement Learning

Earth observation satellite task scheduling research plays a key role in space-based remote sensing services. An effective task scheduling strategy can maximize the utilization of satellite resources and obtain larger objective observation profits. In this paper, inspired by the success of deep reinforcement learning in optimization domains, the deep deterministic policy gradient algorithm is adopted to solve a time-continuous satellite task scheduling problem. Moreover, an improved graph-based minimum clique partition algorithm is proposed for preprocessing in the task clustering phase by considering the maximum task priority and the minimum observation slewing angle under constraint conditions. Experimental simulation results demonstrate that the deep reinforcement learning-based task scheduling method is feasible and performs much better than traditional metaheuristic optimization algorithms, especially in large-scale problems.

Visibility study of Galileo satellites from a VLBI network

<p>Over the last years, ideas have been proposed to install a Very Long Baseline Interferometry (VLBI) transmitter on one or more satellites of the Galileo constellation. Satellites transmitting signals that can be observed by VLBI telescopes provide the opportunity of extending the current VLBI research with observations to geodetic satellites. These observations offer a variety of new possibilities such as high precision tying of space geodetic techniques but also the direct determination of the absolute orientation of the satellite constellation with respect to the International Celestial Reference Frame (ICRF) and have implications on the determination of long-term reference frames. </p><p>This contribution provides a visibility study of the Galileo satellites from a VLBI network. The newly developed satellite scheduling module in VieSched++ is used to determine the time periods during which a satellite is observable from a VLBI network. The possible satellite observations are evaluated through the number of stations from which a satellite is observable. Moreover, the impact on determining the orientation of the satellite constellation, caused by the observation geometry, is investigated with using the UT1-UTC Dilution of Precision (UDOP) factor.</p>