A Coevolution Genetic Algorithm for Conflict Resolution of Multi-user Satellite Observation Activities

With the development of Operationally Responsive Space (ORS), users can directly submit their satellite observation activities to support the emergency events. But it will cause conflicts among these requirements. This paper constructs a graph model for each satellite which maps the satellite observation activities into vertexes. Based the model, the conflict resolution of multi-user satellite observation activities is reduced to a multi-path scheduling problem in graph. The paths are not independent, but have to cooperate with each other to support the users' requirements better. A coevolution genetic algorithm is proposed to solve this problem and a heuristic rule based on Matthew Effect is described in detail to explain how the cooperation information works. A set of experiments is designed to evaluate the effectiveness of the proposed method and the results show that when the conflict is severe, the satisfactory degree of users with high priority can still get better results compared with others.