Abstract:

As China’s space station continues its long-term operations and scientific experiments， the demand for relay satellites has significantly increased， characterized by high frequency， multiple tasks， and diverse services. This complex demand urgently requires more flexible and efficient scheduling solutions for relay satellites to meet personalized service needs of users. This paper proposes an innovative application model for relay satellites， focusing on user preferences and allowing users to submit multiple optional service time windows， as well as specifying the desired execution antennas for each task. To address this new model， we construct a scheduling model for relay satellites that gives a comprehensive consideration of task completion rates， user satisfaction， antenna load balancing， and task priority. We also design a multi-objective scheduling algorithm based on the voting mechanism. This algorithm integrates various multi-objective scheduling methods and adaptively adjusts the weights of these methods during the optimization process， ensuring the selection of the optimal scheduling strategy at different stages. To validate the effectiveness of the proposed model and algorithm， extensive simulation experiments are conducted. The simulation results demonstrate that our method has significant advantages in solving multi-objective scheduling problems for relay satellites， showing remarkable improvements in user satisfaction and system service capacity compared to other multi-objective algorithms such as NSGA-Ⅱ， NSGA-Ⅲ， BiGE， GrEA， MOEA/D， and AMODSA.

Key words: relay satellite, user preference, dynamic scheduling, multi-objective optimization, adaptive