Memetic Multiobjective Discrete Jaya Algorithm for Cooperative Scheduling of Multiple Agile Earth Observation Satellites

• Published in: IEEE transactions on aerospace and electronic systems Vol. 60; no. 6; pp. 8086 - 8099
• Main Authors: Wang, Bin, Feng, Yanxiang, Zhang, Guanghui, Zhang, Lizhi, Yang, Yikang
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.12.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Aerospace and electronic systems; Algorithms; Energy consumption; Integer programming; Linear programming; Load balancing; Local optimization; Mathematical models; Memetics; Mixed integer; Multiple objective analysis; Orbits; Pareto optimum; Satellite observation; Satellites; Task analysis; Task scheduling
• ISSN: 0018-9251; 1557-9603
• DOI: 10.1109/TAES.2024.3429302

Agile earth observation satellites (AEOSs) are widely used in the practical cooperative scheduling of observation tasks. This article addresses a multiobjective AEOS scheduling problem (MO-AEOSSP), which subjects to the energy and memory constraints across multiple AEOSs. The mixed-integer linear programming model of the MO-AEOSSP is formulated. Then, based on the Pareto optimum theory, a memetic multiobjective discrete Jaya (MMOD-Jaya) algorithm is proposed to solve the MO-AEOSSP efficiently. The objectives are to reduce the observation failure rate and enhance the load balancing of energy consumption. Individuals of MMOD-Jaya are generated based on the basic idea of the Jaya algorithm. Two problem-specific objective-improving strategies are developed to reduce the task failure rate and strengthen the load balancing of energy consumption, respectively. They are adopted during the heuristic population initialization and local search scheme. In addition, to prevent from falling into local optimization, a specific self-learning scheme is designed for MMOD-Jaya. Finally, the comprehensive results reveal that the proposed MMOD-Jaya outperforms comparative algorithms in diverse instances.