Research On UAV Dynamic Route Planning With Directional Constraints

With the continuous development of modern technology,the development level of UAV technology has become the symbol of a country's comprehensive national strength.As the core technology of UAV,the route planning technology has attracted wide attention from all over the world.The traditional route planning methods do not consider the direction constraints,once the planning route reach to the target,the plan ended,and there is little research of direction constraints route planning as it requires planning route along a certain direction to get to the target point.With the increasing complexity of the application environment,it is important to study how to safely and accurately approach the target point from a specific direction to improve the penetration ability and combat effectiveness of UAV.This paper has carried out the following research on this problem:Firstly,several key elements that need to be considered in the process of route planning are analyzed,and an equivalent threat model is built based on the types and characteristics of threats in the battlefield environment.When building a planning space model,the equivalent digital map is set up by superimposing the threat into the terrain information in order to reduce the system overhead.Secondly,the principle and characteristic of D* algorithm are analysed,the limitations of the D* algorithm in the route planning problem with direction constraints are pointed out,and analyzes the causes of low efficiency of the algorithm.Finally,the analysis results are verified by simulation experiment.Then,in the route planning with direction constraints,traditional D* algorithm has a long range of planning voyage and high threat cost.To solve these problems,a modified D* algorithm combining genetic algorithm is proposed.The improved algorithm extracts feature nodes from the initial track of D* algorithm planning according to aspecific strategy and generates the initial population based on the optimized feature point path.Then a reasonable fitness evaluation function is proposed,and the optimal route is found by the cyclic iteration of genetic algorithm.The simulation results show that the improved algorithm planning is shorter and the threat cost is smaller,but because the improved algorithm is optimized on the basis of D* programming,the algorithm is more time-consuming than D* algorithm.Finally,a modified D* algorithm combining with the guide points is proposed to solve the problem that the improved D* algorithm with genetic algorithm is time-consuming and can not meet the real-time performance.On the basis of D*,the algorithm introduces the concept of the guide points,and the method of the guide points generation,the selection strategy,and the switching method are studied in detail.Because of generating the guide points,the direction informations of the algorithm nodes extension are increased,and the algorithm planning efficiency is improved.The simulation results show that the improved algorithm can quickly and efficiently plan the feasible route which satisfies the direction constraints in both static and dynamic environment.