Reliable Scheduling Algorithm for Space Debris Monitoring Resources Using Adaptive Multipopulation Differential Evolutionary Optimization With Reinforcement Learning

The continuing growth in space debris has posed a great threat to on-orbit operations. It is urgent to implement reliable and lasting monitoring of space debris. Safety and diversity in monitoring devices and business demands makes scheduling system resources increasingly complicated. This article proposes a novel adaptive multipopulation differential evolutionary algorithm based on a theoretical model specialized in the scheduling of space debris monitoring resources. Using Q-learning, the proposed algorithm adapts self-learning and dynamic adjustment properties in population proportion parameters. Experiments are performed with practical batch tasks and monitoring data to verify the effectiveness and reliable utility of the proposed algorithm to ensure the safety of on-orbit operation.

Automated summary

The continuing growth of space debris poses a significant threat to on-orbit operations. To address this issue, this article proposes a novel adaptive multipopulation differential evolutionary algorithm based on a theoretical model specialized in the scheduling of space debris monitoring resources. Experimental results demonstrate the effectiveness and reliability of the proposed algorithm in ensuring the safety of on-orbit operations.