Agile Earth Observation Satellite Scheduling Algorithm for Emergency Tasks Based on Multiple Strategies
Haiquan Sun,Wei Xia,Zhilong Wang,Xiaoxuan Hu
DOI: https://doi.org/10.1007/s11518-021-5506-4
2021-09-02
Journal of Systems Science and Systems Engineering
Abstract:During the execution of imaging tasks, satellites are often required to observe natural disasters, local wars, and other emergencies, which regularly interferes with the execution of existing schemes. Thus, rapid satellite scheduling is urgently needed. As a new generation of three degree-of-freedom (roll, pitch, and yaw) satellites, agile earth observation satellites (AEOSs) have longer variable-pitch visible time windows for ground targets and are capable of observing at any time within the time windows. Thus, they are very suitable for emergency tasks. However, current task scheduling models and algorithms ignore the time, storage and energy consumed by pitch. Thus, these cannot make full use of the AEOS capabilities to optimize the scheduling for emergency tasks. In this study, we present a fine scheduling model and algorithm to realize the AEOS scheduling for emergency tasks. First, a novel time window division method is proposed to convert a variable-pitch visible time window to multiple fixed-pitch visible time windows. Second, a model that considers flexible pitch and roll capabilities is designed. Finally, a scheduling algorithm based on merging insertion, direct insertion, shifting insertion, deleting insertion, and reinsertion strategies is proposed to solve conflicting problems quickly. To verify the effectiveness of the algorithm, 48 groups of comparative experiments are carried out. The experimental results show that the model and algorithm can improve the emergency task completion efficiency of AEOSs and reduce the disturbance measure of the scheme. Furthermore, the proposed method can support hybrid satellite resource scheduling for emergency tasks.
operations research & management science