电子侦察卫星任务规划调度模型与算法研究

Study of Electronic Reconnaissance Satellites Mission Scheduling Model and Algorithm

对电子侦察卫星任务规划模型及算法进行了研究。首先给出了规划预处理的基本方法，并在合理假设基础上建立起相应的数学模型；然后提出了电子侦察任务可能冲突判断、冲突任务集分割及电子侦察任务冲突执行度等方法及概念，针对遗传算法及模拟退火算法各自的优缺点，设计了一种基于冲突消解的遗传模拟退火算法对问题进行求解；最后通过实例将该算法与普通遗传算法（GA）相比较，结果表明该算法的规划效果比GA更好。

Mission planning for electronic reconnaissance satellites is a challenging combinatorial optimization problem, and its scheduling model and algorithm are studied. First, the problem is analysed and the basic pretreatment method at the front of the planner is given, and then the mathematic models for the problem based on essential hypotheses are establised. Considering the mission requirement constrains and payload constrains, the conceptions of possible conflicts of electronic reconnaissance mission and mission executing conflict degree are presented, and then the whole set of missions are divided into the sets of possible conflicts missions. In order to avoid the disadvantages of GA and SA, a genetic and simulated annealing algorithm based on conflicts resolution is designed for the mission planning of electronic reconnaissance satellites. At last, using an example, the paper shows that the genetic and simulated annealing algorithm＇s performance is much better than GA.