Path Planning Algorithm Of UAV Under The Complex Condition

With the development of air defense technology,unmanned aerial vehicle(UAV)as an important weapon in modern warfare,how to perform a flight mission under complex condition becomes a huge challenge.As the key technology of UAV combat system,path planning is an effective means to improve viability of UAV and make it complete flight mission safety and rapidly.The problem about UAV path planning how to avoid pop-up threat under complex condition is researched in this paper.Firstly,hierarchical path planning strategy is established based on analyzing the requirements of planning time,planning tasks and planning accuracy when UAV proceeding path planning under the different conditions,and the hierarchical path planning method of UAV under pop-up threat is proposed.A*search algorithm is used by the method for global path planning based on the detected battlefield information before UAV takes off,and improved artificial potential field method is adopted for local path planning when UAV detecting pop-up threat,the hierarchical path planning method can ensure UAV back in the initial path after avoiding pop-up threat successfully.Then,a path planning algorithm for UAV under moving threat is proposed in view of the maneuverability of pop-up threat.State information of moving threat is predicted by adaptive kalman filter according to the analysis of maneuvering characteristics of UAV and characters of moving threat,mathematical model for controlling track angle is established based on the relations between heading angle,flight path angle with line-of-sight angles,and path planning algorithm based on parameters adjustment under moving threat is proposed according to the given parameters adjustment regulations in mathematical model for controlling track angle.The simulation results can prove the superiority and effectiveness of the proposed two algorithms.Finally,the research work of this paper is summarized,and the limitation of the research and prospects for future research is pointed out.