The Research Of Unmanned Aerial Vehicles (UAV) Dynamic Path Planning Based On Sparse A* Algorithm And An Evolutionary Algorithm

UAV path planning is an important research content to promote the development of unmanned aerial vehicle(UAV), and dynamic path planning is one of the key technologies to realize the intelligent navigation of unmanned aircraft. With the development of military technology, the flight circumstance of UAV has become increasingly complex. The traditional method of path planning can’t meet the demand of complex battlefield environment of flight. Therefore, it is necessary to develop one kind method which can realize online real-time path planning in the practical complex environment. In order to achieve the real-time dynamic path planning of UAV, this research needs to solve the following four problems:(1) The speed of the path planning;(2) The quality of path planning;(3) For the dynamic real-time flight path planning problem in the case of sudden threat, the dynamic real time flight path planning between the starting point and the fixed target point.(4) For the dynamic real time flight path planning problem in the case of sudden threat, the dynamic real time flight path planning between the starting point and the dynamic target point. In order to solve the problem of UAV dynamic path planning, this paper put forward to a method which is combining the sparse A* algorithm with an evolutionary algorithm, so as to realize the dynamic path planning in complex environment.The main content of this study is including three parts: map model, global static path planning, dynamic real-time path planning.The first part of this research is to establish a map model of flight environment.After analyzing all kinds of map building method, the decision should be made that is to establish a full probability digital map. Then all sorts of predictable threats are regarded as terrains and blended in the known terrain information according to their corresponding threat models, eventually establishing a full probability integrated digital map;The second part of this research is to realize the global static path planning. After the study of existing path planning algorithm, taking into consideration a variety of constraints and mission requirements under the circumstance of the full probability integrated digital map. The sparse A* algorithm and the improved sparse A* algorithm are used to plan a global static track. In order to solve the problem of winding track, it issignificant to realize the static path planning by combining sparse A* algorithm with an evolutionary algorithm, so as to get the optimal static track. Finally using MATLAB software simulation, and compare the three algorithms’ speed and the cost of path.The third part of this research is to realize the dynamic real-time path planning.The global static path is used as a reference track. In the process of actual flight, path planning is selected according to the threat situation and the change of the target. The distance between the emergent threat and the current point of UAV is regarded as the selection criteria. According to the above selection criteria, the priority of the speed of path planning or the optimal solution of the path can be decided, so that the safety of UAV flight can be ensured flexibly. Finally using MATLAB software simulation is an evaluation criterion of the method of the dynamic real-time path planning.The results of MATLAB software simulation show that the method of combined sparse A* algorithm with the evolutionary algorithm solves the problem of winding track in the sparse A* algorithm and realizes the dynamic path planning in the complex environment flexibly, it is significant for the safety and the combat capability of UAV.