面向交通监控的动态无人飞机路径多目标优化  

Multi-objective Optimization of Dynamic Unmanned Aerial Vehicle Cruise Route for Traffic Surveillance

作　　者：王亮[1] 刘晓锋[2] 刘少堂 耿杰[2] 高婷婷[2] 

机构地区：[1]天津市公安交通管理局,天津300040 [2]天津职业技术师范大学汽车与交通学院,天津300222

出　　处：《数学的实践与认识》2017年第22期173-182,共10页Mathematics in Practice and Theory

基　　金：国家自然科学基金(51408417;61503284);国家留学基金委项目([2015]3069)

摘　　要：考虑到无人飞机在交通监控中存在突发的监测目标,需要动态规划无人飞机的巡航路径.首先,引入时间轴的概念,将动态无人飞机路径规划问题分两阶段解决,第一阶段为初始优化,第二阶段为动态时刻点实时优化,进而转化为静态问题.接着,建立了无人飞机路径多目标优化模型,优化目标为无人飞机广义巡航距离最短、无人飞机的使用数量最少.然后,提出了动态无人飞机可行路径插入法,设计了基于帕累托最优的多目标优化算法.最后,进行了案例分析和算法敏感性分析,分析结果表明,提出的模型和算法是可行、有效的.Considering that unmanned aerial vehicle （UAV） may take traffic surveillance for an additional target,it＆#39;s necessary to plan UAV cruise route dynamically.First,the concept of time-axis was introduced,and the dynamical UAV route planning problem was converted into a static problem at two stages.The first stage was initial optimization,and the second stage was real-time optimization at the dynamical time point.Then,a multi-objective optimization model was proposed,which aimed to minimize the UAV generalized cruise distance and minimize the number of UAV used.Next,a dynamical and feasible UAV route insertion method was proposed and Pareto optimality based multi-objective optimization algorithm was designed.Finally,a case study and algorithm sensitivity analysis were conducted,and the analysis result demonstrates that the proposed method and algorithm is feasible and effective.

关 键 词：智能交通 无人飞机 路径优化 交通监控 多目标进化算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] V249[自动化与计算机技术—控制科学与工程] V279[航空宇航科学与技术—飞行器设计]