基于分支剔除的低轨星座实时传感器调度算法  

Real-time sensor scheduling algorithm in LEO constellation based on pruning

作　　者：王博[1,2] 安玮[1] 谢恺[3] 周一宇[1] 

机构地区：[1]国防科学技术大学电子科学与工程学院 [2]中国人民解放军93163部队 [3]解放军炮兵学院军用光电工程教研室

出　　处：《系统工程与电子技术》2010年第6期1244-1250,共7页Systems Engineering and Electronics

基　　金：中国博士后科学基金(20080430223;200801493);安徽省自然科学基金(090412043)资助课题

摘　　要：针对低轨星座目标连续跟踪的传感器资源调度问题,通过对调度约束因素的分析,提取跟踪精度、资源松弛度和资源分配均衡因子三个优化参数,建立了传感器实时调度模型;并建立长时调度决策树,将标准代价搜索方法和分支剔除技术相结合,提出了一种基于分支剔除的实时传感器调度算法。仿真实验表明,标准代价搜索和分支剔除技术的引入明显降低了调度算法的运算量,尤其对于多目标和大步长的情况,且调度传感器跟踪目标的误差略小于短时调度方法。To deal with the sensor resource scheduling in continual object tracking of low earth orbit（LEO） constellation,a real-time sensor scheduling model is established based on three parameters,i.e.,tracking precision,resource slack,and equilibrium factor,according to an analysis of the sensor scheduling constraint factors.Furthermore,a non-myopic scheduling decision tree is developed,and combining the uniform cost search and pruning technology,a novel real-time sensor scheduling algorithm based on pruning is proposed.The simulation results indicate that the calculation load of the sensor scheduling algorithm is reduced obviously,especially to the cases of multi-target and longer step length,moreover,the tracking error of the proposed method is slightly lower than the myopic one.

关 键 词：低轨星座 调度算法 分支剔除 模型 跟踪 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置]