Multi-satellites imaging scheduling using individual reconfiguration based integer coding genetic algorithm

• Published in: Acta astronautica Vol. 178; pp. 645 - 657
• Main Authors: E., Zhibo, Shi, Renhe, Gan, Lan, Baoyin, Hexi, Li, Junfeng
• Format: Journal Article
• Language: English
• Published: Elmsford Elsevier Ltd 01.01.2021; Elsevier BV
• Subjects: Algorithms; Combinatorial analysis; Computing costs; Earth observations (from space); Genetic algorithm; Genetic algorithms; Individual reconfiguration; Integer programming; Multiple satellites scheduling; Optimization; Reconfiguration; Satellite imagery; Satellite observation; Satellites; Scheduling; Windows (intervals); Individual reconfiguration; Genetic algorithm; Multiple satellites scheduling
• ISSN: 0094-5765; 1879-2030
• DOI: 10.1016/j.actaastro.2020.08.041

Scheduling for the Earth observation satellite (EOS) imaging mission is generally considered as a complicated combinatorial optimization problem subjected to various technical constraints, which requires massive computational costs to find the optimal solution, especially for multiple EOSs imaging missions. In this paper, a novel individual reconfiguration based integer coding genetic algorithm (IRICGA) is developed to reduce the computational costs and improve the optimality of multiple EOSs scheduling for area target observation. The proposed individual reconfiguration procedure contributes to generating feasible solutions during the evolutionary process. Considering the diversity of individual population, two different reconfiguration mechanisms are proposed for handling various technique constraints. Based on the proposed algorithm, an efficient multi-satellite imaging scheduling framework for area target observation is developed. The scheduling framework consists of two separate phases, i.e., pro-processing and scheduling process. In the pro-processing phase, a semi-analytical method is proposed to calculate the visible time window (VTW) of area target and observation strip. Moreover, the binary search techniques are utilized to improve calculation efficiency and accuracy. Besides, a new area partitioning method based on two kinds of discrete parameters is proposed to divide the area target into a series of feasible observation strips. Based on the pro-processing results, the multiple EOSs scheduling problem is formulated as an integer programming model. In the scheduling process, based on the analysis of characteristics of the multiple EOSs scheduling problem, the IRICGA is constructed to generate the optimal scheduling solution. In the end, a real-world multiple EOSs scheduling example is investigated to illustrate the high-efficiency and reliability of the proposed method. •Individual reconfiguration mechanisms are developed to generate feasible solutions.•An IRICGA is proposed to solve the multi-satellite imaging scheduling problem.•An efficient multi-satellite scheduling framework is developed based on the IRICGA.•Effective techniques are proposed to decrease the calculation cost of satellite scheduling.•The simulation results reveal that the proposed algorithms display high-efficiency.