基于切比雪夫分解算法的实时无人飞机交通监控巡航路径规划  

Real-time drone cruise route planning to monitor traffic situation using the Tchebycheff decomposition-based algorithm

作　　者：陈强 刘晓锋[1] 刘军黎 牛树云[2] Chen Qiang;Liu Xiaofeng;Liu Junli;Niu Shuyun(School of Automotive and Transportation,Tianjin University of Technology and Education,Tianjin 300222,China;ITS Center,Research Institute of Highway Ministry of Transport,Beijing 100088,China)

机构地区：[1]天津职业技术师范大学汽车与交通学院,天津300222 [2]交通运输部公路科学研究院ITS中心,北京100088

出　　处：《国外电子测量技术》2023年第5期142-149,共8页Foreign Electronic Measurement Technology

基　　金：天津市智能交通技术创新团队重点培养专项(XC202028);天津市教委科研计划项目(2021KJ018);天津市科技计划项目(2022ZD016,22YDTPJC00120)资助。

摘　　要：将无人飞机技术引入到道路监控当中,以提高交通监控的灵活性和响应速度。针对道路突发交通事故、路段坍塌、交通拥挤等情形,交通监控目标的数量和分布会发生动态变化,此时需要协同地面车辆和无人飞机的监控分工关系,动态规划无人飞机巡航路径。首先,建立了无人飞机交通监控巡航路径多目标优化模型,该模型考虑了无人飞机巡航与地面车辆巡查的协同关系以及监控成本,优化目标包括无人飞机-车辆的协调性最好、无人飞机的巡航距离最小。接着,设计了基于分解策略的多目标优化算法,该算法包括切比雪夫分解策略、无人飞机可行路径染色体的生成、子路径的划分、交叉、变异等操作,采用帕累托最优确定最优解。最后,开展了案例分析。结果表明,与3种多目标优化算法相比,提出的算法可以求出无人飞机巡航路径的最优解,算法的时间复杂度显著降低,平均计算时间下降了50%以上,解的质量也有一定程度的提升,这表明提出的方法可用于实时无人飞机交通监控巡航路径优化研究。Drone technology was introduced to monitor traffic situation,so as to improve the flexibility and response speed of traffic monitoring.Considering sudden traffic accident,road collapse and traffic congestion,the number and distribution of targets will change dynamically,thus it was necessary to coordinate the tasks of ground vehicle and drones and plan drone cruise routes in real-time.First,a multi-objective optimization model was proposed,which considered the collaborative relationship between drones and vehicle,and the drone cruise cost.The optimization objectives aimed to maximize the coordination between drones and vehicle,and minimize the drone cruise distance.Then,a decompositionbased multi-objective optimization algorithm was designed,which included the Tchebycheff decomposition strategy,the feasible drone route chromosome generation,sub-route division,crossover and mutation operation,and the Pareto optimality was adopted to obtain the best solution.Finally,a case study was implemented and the analysis results show that compared with three multi-objective optimization algorithms,the proposed algorithm can obtain the best solution of drone cruise routes;in addition,the time complexity of the proposed algorithm decreases significantly,and the average calculation time decreases by more than 5o%,and the solution quality has been improved.This demonstrates that the proposed model and algorithm can be used to plan the real-time drone cruise routes for traffic monitoring.

关 键 词：智能交通 路径规划 无人飞机 多目标优化 交通监控 

分 类 号：U491[交通运输工程—交通运输规划与管理]