求解卫星地面站调度问题的演化学习型蚁群算法

Learnable ant colony optimization algorithm for solving satellite ground station scheduling problems

随着观测需求的日益增加,越来越多的卫星和地面站加入到对地观测系统中,迫切需要采用科学手段对卫星地面站资源进行合理分配。针对卫星地面站调度问题,构建了一种演化学习型蚁群算法。实验结果表明,该算法能有效求解卫星地面站调度问题。将蚁群优化模型和知识模型进行优势互补,可极大提高演化学习型蚁群算法的效率,为现有优化方法的改进提供了一种有益借鉴。

With the increased observing requirements, more and more satellites and ground stations are joined to the earth observing system. It is urgent to effectively allocate the satellite ground station resources using some scientific techniques. Aiming to the satellite ground station scheduling problem, a learnable ant colony optimization （LACO） algorithm is proposed. Experimental results show that LACO is a viable and effective approach for the satellite ground station scheduling problem. This approach legitimately combines the ant colony optimization model with the knowledge model, which largely pursues the integrating advantages of these models. The proposed approach orovides a useful reference to the improvement of existing optimization approaches.