基于协同进化的多平台联合对地观测优化调度

Scheduling multi-platforms collaborative observation based on cooperative evolution

在分析了卫星与无人机在执行观测与资源调度上的特性差异基础上,建立了多平台联合对地观测调度问题的数学模型,提出了多平台协同进化调度算法(MPCCPSA)进行求解。MPCCPSA采用分层式协同进化架构解决了不同类型观测方案统一调度生成问题。根据不同类型平台使用特性以及观测目标集合特点,采用分治-合作策略将其分解分配到各平台,顶层的交叉、变异操作保证各种群的多样性,底层的分治、合作算子保证卫星与无人机之间保持观测能力动态互补,在确保可行解的前提下加快收敛速度。仿真实验表明该方法能够有效解决空-天基多类型平台联合观测优化调度问题。

Based on the analysis of the features and differences between satellite and UAV in earth observation and scheduling, a mathematical model was presented to formulate the scheduling problem for the multi-platforms collaborative observation, and a multi-platforms cooperative evolutionary planning and scheduling algorithm （MPCCPSA） was proposed to solve this problem. MPCCPSA uses co-evolution framework for solving the different platforms＇observing plans generated in a unified manner. Based on the divide-cooperation strategy and different characteristics of platforms, the observed targets were allocated to each platform. On the top-level, crossover and mutation operations ensured the diversity of the various groups, and the underlying divide-cooperation operators ensured the dynamic complementarity between different platforms. Simulation results show that this method can effectively solve the multi space-aeronautics collaborative observation tasks scheduling problem.