Journal of System Simulation ›› 2025, Vol. 37 ›› Issue (7): 1836-1847.doi: 10.16182/j.issn1004731x.joss.25-0095

• Invited Papers • Previous Articles    

Multi-scenario Multi-satellite Mission Planning Method Based on Adaptive Large Neighborhood Search

Li Xiutian1, Wang Ling2, Chen Yingwu1, Xing Lining3, Chen Yingguo1   

  1. 1.College of Systems Engineering, National University of Defense Technology, Changsha 410073, China
    2.Department of Automation, Tsinghua University, Beijing 100084, China
    3.Research Institute of Intelligent Control and Manufacturing System, Jiangsu University of Technology, Changzhou 213001, China
  • Received:2025-02-10 Revised:2025-04-13 Online:2025-07-18 Published:2025-07-30
  • Contact: Xing Lining

Abstract:

To further improve the execution efficiency of remote sensing satellites, an integrated optimization framework combining adaptive large neighborhood search (ALNS) and a constraint programming-boolean satisfiability problem (CP-SAT) solver monitor was proposed, addressing the challenges of complex constraints, dynamic scale, and resource heterogeneity in multi-scenario multi-satellite mission planning. A unified multi-objective mixed-integer programming model was established, coupling heterogeneous constraints of point targets and area tasks. A time-domain rolling mechanism dynamically decomposed the problem scale, and a priority screening strategy enhanced the search efficiency of ALNS. Solution feasibility was verified in real time through the CP-SAT monitor. Results show that compared with genetic algorithm, particle swarm optimization, and deep Q-network, the proposed method achieves a 15%~28% improvement in task completion rate, 30%~50% reduction in computation time, and over 20% optimization in load balancing in 300 test scenarios.

Key words: satellite mission planning, multi-scenario, adaptive large neighborhood search, mixed-integer programming, dynamic priority screening

CLC Number: