Research And Development On Route Planning In UAV’s Tasks Of Reconnaissance And Air-to-ground Attack

In recent years, Unmanned Aerial Vehicles (UAVs) are widely used in local warfares and play anirreplaceable role. Route planning is one of the critical technologies of UAVs’ autonomous flight. Asprevious study mostly focuses on the way to plan safer and more fuel-efficient routes which are usedfor low altitude penetration, this thesis will respectively research the methods of route planning indifferent flight phases of reconnaissance task and air-to-ground attack task.Through the analysis of reconnaissance process and air-to-ground attack process, the thesisdivides these two processes into such five flight phases as stealthy penetrating, fixed reconnaissance,target seeking, target tracking and target attacking. In each phase, UAV’s flight route needs to meetsome specific requirements. In order to describe the requirements clearly and accurately, and then toresearch appropriate methods of route planning based on the requirements, the thesis firstlysummarizes three elements that should be considered when planning routes in all phases, includingenvironmental element, UAV system element and target element. The elements’ mathematical modelsare also built in detail.With the models of route planning elements, this thesis studies the methods of route planning inthe five flight phases. The offline and online planning methods of penetration route are designedbased on sparse A-star algorithm and D-star algorithm respectively; If a fixed reconnaissance mode isdesignated, then UAV’s reconnaissance route can be optimized only by optimizing some relatedparameters so as to shorten the time of flight, reduce the number of photos, and improve the efficiencyof airborne sensor’s coverage; When the UAV is searching or tracking targets, the route planningproblems are both described as rolling horizon control(RHC) problems, and respectively solved byintroducing the particle swarm algorithm(PSA) and genetic evolution algorithm(GEA), thus the UAVcan find targets as many as possible within a short period of time, and track targets with access tomore certain state information of targets but low costs; The route planning method in target attackingphase is an improved method on the basis of the proportional guidance law, and the route planed isable to guide the UAV to enter the accessible zone of airborne air-to-ground missile quickly andaccurately, so as to fulfill the task of precisely attacking targets on the ground.With the above route planning methods of each flight phase as the theoretical basis, the thesisuses C programming language to develop a software system specialized for route planning in theprocesses of reconnaissance and air-to-ground attack. This system is composed of planning elements module, route planning module and software interface module. First of all, every module’s I/Ointerfaces and actions are determined by using modular design method, and then, through writingcodes the interfaces and actions can be turned into the data and functions in procedures, after that, thesoftware needs to be debugged and tested. At last, the test results show that this system has realizedthe route planning for UAV’s implementation of reconnaissance task and air-to-ground attack task.