In order to resolve the scheduling difficulty in the control of command system,the resource scheduling problem of a special sensor was studied and the constraint conditions including detected times,the interval between two detections,and the resource restrict of sensor were analyzed.A 0-1 programming model was established and transformed to a 0-1 liner integer model whose objective function is the sum of the importance degree of tracking targets.The optimal solution which can reasonably schedule sensor resource when the workload is saturated was obtained by using the cutting plane algorithm.A corresponding simulated annealing algorithm was proposed to improve the speed of solving and was used to solve some examples.Compared with the cutting plane algorithm and the genetic algorithm in terms of resource utilization and the speed of solving,the validity of the proposed model and the high efficiency of the simulated annealing algorithm were proved.%为解决指挥系统控制中的调度困难,研究了一类特殊的传感器资源调度问.主要分析了跟踪目标的探测次数、时间间隔和传感器资源等约束条件.用跟踪目标的重要程度之和作为目标函数,建立了一个0-1规划的数学模型,再利用变换将其转化为0-1线性整数规划模型.利用割平面法求解得出最优调度策略,其能在工作量饱和的情况下合理调度传感器资源.为提高求解速度,提出了对应的模拟退火算法.通过对一些不同规模实例的求解,在资源利用率和算法的求解速度等指标上,与割平面法及遗传算法进行对比分析,验证了模型的有效性和模拟退火算法求解的高效性.
展开▼
