大地震灾区影像拍摄任务规划与跨平台对地观测系统多资源应急调度优化

It is crucial to obtain quickly disaster condition for a successful rescue after a high-magnitude earthquake. Thus, it becomes quite important to plan photograph for the disaster area and schedule resources distributed in the different observation systems. Firstly, we will analyze disaster geographical characteristics and the performances of observation resources, and then propose a networking-decomposition-based method for decoupling the disaster observation mission. We will build the task assignment model and then exploit its properties. By these properties, we develop the heuristics for scheduling observation resources. The computational complexity will be investigated. Finally, we will reconfigure some typical cases through well-known high-magnitude earthquake, and then explore the fast rule of photograph mission decoupling and the optimal cooperation mode of observation resources in these different cases, in order to enhance the practicality of the proposed models and algorithms. The results of this project can improve the existing emergency management theory, but also help the government to rescue promptly the victims.

在大地震发生后，即时获取灾情是成功救援的前提与关键。研究灾区影像拍摄任务规划与跨平台对地观测系统多资源应急调度问题对即时获取灾情有重要的指导意义。首先，根据灾区地理环境信息和观测资源性能，研究基于灾区网格化分解的拍摄任务解耦方法。然后，构建大地震灾区拍摄任务分配模型并分析其解的特性，探索启发式求解规则，开发多资源应急调度观测的优化算法并分析其计算复杂性。最后，参考一些知名的大地震实例，重构典型的应用情景，挖掘不同情景下的拍摄任务快速解耦规则和跨平台多资源应急调度最佳协同模式，提升所建模型和所提算法的工程实用性。研究结果不仅可以完善现有应急管理理论而且还有助于提升政府地震应急管理的水平。

在大地震发生后，即时获取清晰的灾区影像是组织灾后精准救援工作的关键。针对灾区影像拍摄任务规划与跨平台对地观测系统多资源应急调度问题，提出了基于灾区网格化分解的拍摄任务解耦方法，构建了大地震灾区拍摄任务分配模型，设计并开发了多资源应急调度观测的智能优化算法，对比分析了不同启发式求解算法的计算效率。研究成果丰富了卫星调度和智能优化理论，为政府快速做出地震应急救援工作部署提供了决策基础。..在项目执行期内，已公开发表论文13篇，其中发表在IEEE Trans、 EJOR、IJPR、系统工程理论与实践等国内外高水平期刊上的论文12篇，总引用340次，单篇最高引用52次。依托该项目，获得软件著作权2项，培养研究生3名。