Dynamic Path Planning Of UAV Based On Three-Dimensional Geographic Scene

With the wide application of UAV in various fields and the increasing complexity of UAV flight environment,there are more challenges and requirements for UAV path planning technology.At present,scholars and experts at home and abroad have carried out fruitful research on UAV path planning methods,but there are still some shortcomings in the threedimensional complex constraint environment.And the methods are only suitable for a specific environment and its' adaptability is insufficient.Therefore,in this paper,a multi-layer extended A * algorithm is designed for 3D global static path planning of UAV,so that the planning can take into account the complex environmental constraints and accord with the flight characteristics of UAV.In the problem of 3D local dynamic path planning for UAV,the traditional particle swarm optimization algorithm is improved,and the optimal guidance point is searched to modify the path to avoid the sudden threat and improve the convergence efficiency of the algorithm.Finally,the comparative analysis experiments of UAV global static path planning and local dynamic path planning are carried out,and the main research work and results are as follows:(1)A multi-layer extended A* algorithm for 3D global static path planning of UAV under complex environment constraints is designed.Before the algorithm search,a typical complex environment(terrain,weather,no-fly zone,dangerous object,et al.)is fused with a track constraint condition under different task targets,and a three-dimensional path planning environment model is constructed to provide a search environment for the algorithm,so as to adapt to the diversified constraint requirements under different environments.Considering the flight characteristics and performance constraints of UAV,a multi-layer extended A * algorithm is designed to improve the accuracy of the path through the "new horizontal direction vector" and the "vertical hierarchical expansion node".The flight cost is normalized to improve the cost function of the algorithm to guide the selection of the optimal reference path,which makes the algorithm more efficient.(2)A three-dimensional Local dynamic path planning method for UAV based on optimal guidance point and the checking method of path with equal grid are proposed.In order to solve the problems of slow efficiency and low accuracy of 3D local dynamic path planning method for UAV,after obtaining the local dynamic planning area,In order to avoid the sudden threat,the rapid planning can be realized by directly finding the optimal guiding point to modify the path in the local range.When finding the optimal guidance point,the strategy of "hierarchical random initialization" and "particle lazy release" is used to improve the convergence efficiency of particle swarm optimization algorithm.In order to verify the correctness of the planned threedimensional path,a new checking method of path grid is proposed,which is divided into a series of nodes by grid-by-grid,and then checked one by one in order to verify the correctness of the planned three-dimensional path.(3)Develop the prototype system and carry on the experiment analysis.Taking the terrain undulating area near Jinsha River in Ganzi Tibetan Autonomous Prefecture of Sichuan Province as a case area,firstly,the global static path planning experiment of UAV is carried out through the multi-layer extended A* algorithm proposed in this paper,and compared with the classical A* algorithm.Sparse A * algorithm is compared and analyzed to verify the efficiency and accuracy of the algorithm proposed in this paper.Then the sudden threat is added,and the local dynamic path planning experiment is carried out through the method based on the optimal guidance point proposed in this paper.The improved algorithm and the basic particle swarm optimization algorithm are compared and analyzed to verify the efficiency of the improved algorithm.Finally,the planned path are checked to verify the accuracy and applicability of the proposed method.