Coordinated Scheduling of Air and Space Observation Resources via Divide-and-Conquer Framework and Iterative Optimization

At present, independent scheduling of Earth-observation resources (EORs) is usually difficult to satisfy diverse observation requirements and cannot realize the full potential of space–air resource networks. To utilize EORs comprehensively, this study constructs a divide-and-conquer framework (DCF) for a coordinated scheduling of air and space observation resources (i.e., satellites and unmanned aerial vehicles). The DCF can decompose the original scheduling problem into a task allocation subproblem and multiple task scheduling subproblems that can be solved using a coordination planner and subplanners, respectively. For the task allocation subproblem, we propose a simulated annealing algorithm combined with variable neighborhood adjustment (SA-VNA) method, where a solution variation strategy (SVS) is designed. The SVS iteratively adjusts the task allocation scheme according to the coordinated scheduling result of the last iteration. Based on the allocation scheme, multiple task scheduling subproblems are generated, and existing effective algorithms are used to resolve them. Extensive experiments and comparative analysis show that SA-VNA outperforms several peer algorithms overall, indicating that DCF plus SA-VNA can significantly improve the efficiency of space–air resource networks.