Rapid Clustering and Cleanup Method for Space Debris Based on Improved K-Means

Space debris poses a growing threat to satellite operations and space missions, necessitating efficient removal strategies. This study proposes an improved K-Means clustering algorithm that incorporates weighted orbital parameters and a secondary refinement step to optimize debris grouping. Combined with a genetic algorithm for rapid transfer cost estimation and mixed integer/continuous optimization, the approach ensures adherence to fuel consumption constraints while enhancing clustering efficiency. Simulations on three test cases involving 21, 60, and 120 debris objects demonstrate significant performance improvements. The improved K-Means algorithm achieves up to 81.07 % fuel savings compared to traditional methods. The results highlight the method's robustness, scalability, and potential to enhance the efficiency of space debris removal, contributing to long-term space sustainability and operational safety.