Predictive-Reactive Scheduling for Space Missions in Small Satellite Clusters

Task scheduling in small satellites clusters is an important optimization problem because of limited resources and various constraints. Specifically, automated and dynamic scheduling is necessary for practical schedule and real-time response of exceptions, whereas it makes the problem more complicated. In this paper, a predictive-reactive scheduling model has been proposed for space missions in heterogeneous small satellite clusters, which is realized by the hybrid of periodic task scheduling and dynamic adjustment. In the model, the characteristics and constraints of various space missions are abstracted and then multi-types of tasks are defined from different perspectives. Additionally, the link status is predicted and introduced as the communication constraint in the scheduling process, due to the frequent topology changes and the unreliable communication links. A priority-based scheduling algorithm ELSFT has been proposed and applied in the scheduling process. By compared with both classis greedy algorithms and the latest heuristic search algorithm, the experimental results show that our approach provides real-time schedules with high completion rate and minimum makespan.