A clustering-reconstruction-based multi-objective refueling mission scheduling framework for large-scale geosynchronous orbit spacecraft

In the contemporary landscape of burgeoning space exploration activities, on orbit refueling (OOR) missions play a pivotal role in the long-term sustainability and expansion of space operations. However, as the scale of the demand for on orbit service (OOS) missions grows in tandem with the increasing number of spacecraft launched into space, the existing mission scheduling strategies struggle to meet the requirements of large-scale OOR. To cope with this challenge, this paper presents a Clustering-Reconstruction-Based multi-objective refueling mission scheduling framework for large-scale spacecraft. In this proposed framework, an Exact Euclidean Locality Sensitive Hashing based on Cosine Similarity Selecting (E2LSH-CSS) clustering method is employed to implement grouping target spacecraft, which effectively reduces the complexity of the large-scale scheduling problem by decomposing it into multiple sub-problems. This method applies E2LSH to construct a hash table containing a wide array of target spacecraft groupings, then selects the optimal target grouping according to silhouette analysis based on cosine similarity. Within each sub-problem, the Non-dominated Sorting Genetic Algorithm II (NSGA-II) is introduced to generate optimal Pareto fronts in each sub-problem. Furthermore, a Multi-Objective Reconstruction Algorithm (MORA) is used to reorganize the sub-problems to obtain the global optimal Pareto front. Finally, extensive numerical simulations demonstrate the effectiveness of the proposed scheduling framework, showcaseing its flexibility and scalability in obtaining optimal mission scheduling schemes.