The paper investigates the emergency scheduling problem of multiple aerial targets staring surveillance with multiple optical satellites (MSMT). To fulfill the requirement of state estimation and trajectory prediction, a novel scheduling model including specified duration of cooperative observation and revisits has been proposed. The total observation profit, introducing the independent observation for detection efficiency and cooperative observation for positioning precision, is regarded as the optimization objective. The corresponding time-varying observation gain functions of the two kinds of observations are illustrated respectively. Combined with the constraints of visibility, optic axis (OA) adjustment ability, cooperation duration and immunity, the constraint satisfaction mission scheduling model is established. In addition, a cooperation-oriented ant colony optimization algorithm (CO-ACO) is designed to determine the observation sequence for each satellite. Five elements for heuristics, balancing the quality and quantity of observations, are introduced to the design which can fully exploit the intelligence and improve the algorithm effectiveness. Simulations are performed to verify the effectiveness and demonstrate the superiority to existing algorithms under different conditions in terms of optimization performance.

本文研究了利用多颗光学卫星（MSMT）监视多个空中目标的紧急调度问题。为了满足状态估计和轨迹预测的要求，提出了一种新的调度模型，该模型包括指定的合作观察持续时间和重访时间。引入观测效率的独立观测和定位精度的协同观测的总观测利润被视为优化目标。分别说明了两种观测的相应时变观测增益函数。结合能见度，光轴（OA）调整能力，协作持续时间和免疫力等约束条件，建立约束满足任务调度模型。此外，设计了面向合作的蚁群优化算法（CO-ACO），以确定每个卫星的观测序列。在设计中引入了五个启发式元素，平衡了观察的质量和数量，可以充分利用智能并提高算法的有效性。进行仿真以验证有效性，并在优化性能方面证明其在不同条件下相对于现有算法的优越性。