HIGH TECHNOLOGY LETYERSI V01．17 No．I Mar．2011 PP．13一18doi：10．3772／j．issn．1006-6748．201 1．01．003Three·dimension path planning for UAV using improvedA木algorithm in complicated threat environment①Li Xia(李霞)②’，Wei Ruixuan’，Wang Zhike”(+Institute of Zngineefing，Air Force Engineering University，Xi’an 710038，P．R．China)(”Air Force Aviation University，Changchun 130022，P．R．China)AbstractIn order to impmve the battle effectiveness of the unmanned aerial vehicle(UAV)under the in-creasingly complex threat environment，a three-dimension path planning method based on an A‘al-gorithm is proposed and simulated inthis paper which makes the UAV satisfy requirements ofdiffer-ent missions．At first，the digital map information is processed with an integrated terrain smoothingalgorithm．and a safe flight surface which integrates the vehicle dynamic i8 built and added on theterrain，and tllen，models of the complicated threats are established and integrated into tlle digitalterrain．At last，an improved A+algorithm is used to plan the three-dimension path on the safe sur-face，and then smooth the path．Simulation results indicate that the approach has a good perform-alice in creating an optimal path in the three—dimension environment and the path planning algorithmis more simple，efficient and easily realized in the engineering field．Key words：improved A+algorithm，path planning，digital map，path smoothing0 IntroductionUnmanned aerial vehicle(UAV)is an inevitabletrend of the modern aerial weapon equipment．Re-search on UAV affects battle effectiveness of the airforce and is important for the safeness of a nation．Pathplanning is to generate a space path between an initiallocation and a desired destination that has an optimalor near-optimal performance under specific constraintconditions，and it is an imperative task required in thedesign of UAV?．Finding an optimal path planning isone of themost applicable problems，especially in robot industry，military and etc．[2J．The flight path planning in a largemission area is a typical large scale optimization prob-lem，a series of algorithms have been proposed to solvethis complicated multi--constrained optimizationprob·-lem．In Ref“2]，a new approach based on the meas-ure theory for finding an approximate optimal path inthe presence of obstacles is presented．In Ref．[3]，anapplicable method for solving the shortest path is pro-posed．Ref．[4]presents a dynamic optimalcontrolmethod for autonomous trajectory planning and control·ling an Unmanned Ground Vehicle(UGV)using real-time information updates，the objective of the UGV isto traverse from an initial start point and reach its goalin minimum time with maximum robustness．Ref．[5]presents a new approach for solving optimal path plan-ning problems for single rigid and free moving object intwo and three dimensional space in the presence of sta-tionary or moving obstacles．In all of the above refer-ences，the threats model is very simple．And there脚呛a lot of other methods for path planning，such as evolu·tionary computation[6】and genetic algorithm【7]whichcost too much time，and the methods can hardly solvethe contradiction between the global optimization andthe complicated threats environment．Furthermore，thecurrent work mainly focuses on two dimensions．Combining the traditional A‘algorithm，the im-proved one is proposed in this paper，which can over-come the contradiction between the global optimizationand the complicated threats environment ofe】【isfingpath planning algorithm for UAV and is used to planthe three．dimension path with better speed．Our meth．odcan reduce the planning space efficiently and im-prove the search efficiency of the algorithm，and thesimulation has validated this algorithm．This paper includesfour sections．Section l con．structs a planning space which guarantees the fli【ght so-curity and the mathematicalmodels of the complicatedthreats are established and integrated into the digital①Suppofled by the National HigII Technology Research and Development Programme of China②To whom correspondence should be addressed．E—mail：li】【iakgd2010@163．cornReceived on Jan．25，2010万方数据
