多类异构对地观测平台协同任务规划方法

目前,不同类型的对地观测平台之间缺乏有效的协同交互机制。这种孤立的资源管控模式难以应对多样且大量的对地观测需求。特别是在一些紧急情况下,如地震、武装冲突、洪涝灾害和森林火灾等,这种模式的弊端尤为突出。研究了多类异构观测资源,包括卫星、飞艇及无人机(UAV)的协同规划问题。首先,提出一种基于多Agent的分层协同规划框架,整合不同观测资源构成一个分布式和松耦合的对地观测系统。其次,将异构对地观测平台的协同规划问题转化为不同子规划中心间的任务分配问题。第三,针对该任务分配问题,提出一种结合禁忌列表模拟退火(SA-TL)算法,在该算法中融合了禁忌表策略,有效提高了算法的性能。仿真实验验证了多Agent协同框架的优越性和SA-TL算法的效率。

Coordinated task planning method of multiple heterogeneous Earth-observation platforms

Currently, different Earth-observation resources are usually managed by different organization sub-planners which lack interactions and cooperation among each other. Such independent resource operations are no longer efficient to meet diverse and vast observation requests, especially in emergencies, such as earthquakes, flooding and forest fire disasters. This paper addresses the issue of the coordinated planning of heterogeneous observation platforms, including satellite, airship and unmanned aerial vehicle(UAV). First, an agent-based coordinated planning architecture is proposed to integrate heterogeneous observation resources into a distributed and loosely-coupled earth-observation system. Second, the coordinated task planning problem is transformed into a problem of the task assignment among sub-planners. Third, a simulated annealing algorithm combined with a tabu list(SA-TL) is proposed to solve the task assignment problem. SA-TL is integrated with a tabu-list strategy, which improves the performance efficiently. Experiments and comparative studies demonstrate the superiority of agent-based architecture and the efficiency of SA-TL.