Learning-guided nondominated sorting genetic algorithm II for multi-objective satellite range scheduling problem

Abstract

Abstract Satellite range scheduling is an important issue in the field of satellite mission planning, which greatly affects the development of satellite industry. This paper analyzed satellite range scheduling problem, constructed a multi-objective SRSP (MO-SRSP) model and proposed an improved multi-objective evolutionary algorithm (MOEA), called learning-guided nondominated sorting genetic algorithm II (LGNSGAII) that contains a learning mechanism. Learning mechanisms can speed up optimization process. Meanwhile, another algorithm called task-time window selection algorithm (TTSA) is also proposed. Specifically, it can select satellite ground stations time windows for tasks. TTSA includes three location selection methods and two location movement methods, both are used to select the appropriate execution location for tasks. Experiments show that the proposed algorithm can solve MO-SRSP better than several comparison algorithms. In other words, this algorithm we proposed has a broad practical application prospect.