Stecz Wojciech, Gromada Krzysztof
Faculty of Cybernetics, Military University of Technology, 00-908 Warsaw, Poland.
PIT-RADWAR, 04-051 Warsaw, Poland.
Sensors (Basel). 2020 Feb 17;20(4):1080. doi: 10.3390/s20041080.
The paper presents the concept of mission planning for a short-range tactical class Unmanned Aerial Vehicle (UAV) that recognizes targets using the sensors it has been equipped with. Tasks carried out by such systems are mainly associated with aerial reconnaissance employing Electro Optical (EO)/Near Infra-Red (NIR) heads, Synthetic Aperture Radar (SAR), and Electronic Intelligence (ELINT) systems. UAVs of this class are most often used in NATO armies to support artillery actions, etc. The key task, carried out during their activities, is to plan a reconnaissance mission in which the flight route will be determined that optimally uses the sensors' capabilities. The paper describes the scenario of determining the mission plan and, in particular, the UAV flight routes to which the recognition targets are assigned. The problem was decomposed into several subproblems: assigning reconnaissance tasks to UAVs with choosing the reconnaissance sensors and designating an initial UAV flight plan. The last step is planning a detailed flight route taking into account the time constraints imposed on recognition and the characteristics of the reconnaissance sensors. The final step is to generate the real UAV flight trajectory based on its technical parameters. The algorithm for determining exact flight routes for the indicated reconnaissance purposes was also discussed, taking into account the presence of enemy troops and available air corridors. The task scheduling algorithm-Vehicle Route Planning with Time Window (VRPTW)-using time windows is formulated in the form of the Mixed Integer Linear Problem (MILP). The MILP formulation was used to solve the UAV flight route planning task. The algorithm can be used both when planning individual UAV missions and UAV groups cooperating together. The approach presented is a practical way of establishing mission plans implemented in real unmanned systems.
本文提出了一种针对短程战术级无人机（UAV）的任务规划概念，该无人机利用其配备的传感器识别目标。此类系统执行的任务主要与采用光电（EO）/近红外（NIR）机头、合成孔径雷达（SAR）和电子情报（ELINT）系统的空中侦察相关。这类无人机在北约军队中最常用于支援炮兵行动等。在其活动期间执行的关键任务是规划侦察任务，确定能最佳利用传感器能力的飞行路线。本文描述了确定任务计划的方案，特别是分配识别目标的无人机飞行路线。该问题被分解为几个子问题：为无人机分配侦察任务并选择侦察传感器，以及指定无人机初始飞行计划。最后一步是考虑识别任务的时间限制和侦察传感器的特性来规划详细的飞行路线。最后一步是根据无人机的技术参数生成实际的飞行轨迹。还讨论了考虑敌方部队存在和可用空中走廊来确定指定侦察目的的精确飞行路线的算法。任务调度算法——带时间窗的车辆路径规划（VRPTW）——以混合整数线性问题（MILP）的形式制定。MILP公式用于解决无人机飞行路线规划任务。该算法可用于规划单个无人机任务以及协同作业的无人机群。所提出的方法是在实际无人系统中制定任务计划的一种实用方式。
