Research On Obstacle Avoidance Flight Path Algorithm Of Forest Fire UAV

With the booming development of UAV technology,the application of UAV in the field of forest fire is becoming more and more extensive.However,the forest environment is complex,the environmental information obtained by UAV before the flight mission may not be comprehensive,and the UAV may encounter sudden threats during the flight.Therefore,it is of great significance to study the global flight path planning of UAV and local flight path planning of sudden threat.This paper adopts the idea of hierarchical planning to study the global flight path planning and local obstacle avoidance flight path planning algorithms of forest fire fighting UAV respectively.The specific contents are as follows:(1)Overall flight path planning of UAV.Firstly,three dimensional digital map was used to simulate the terrain of mountain forest to build a track planning model.Then,aiming at the shortcomings of Wolf pack algorithm in UAV global flight path planning,Circle chaotic mapping was used to improve the population initialization mode,so as to improve the quality and diversity of population initial solution.To improve the global search capability of Wolf pack algorithm,the fixed step size of wandering behavior is changed to Levi’s flight step size.Set the limit of rush times,and change the fixed step size of calling behavior and siege behavior to random step size to prevent the algorithm from falling into local optimal.Finally,the improved Wolf pack algorithm is used to search the global track in the simulated mountain terrain environment,and the comparison test results show that the improved Wolf pack algorithm can plan a shorter track and the planned final track length is more stable.(2)Local obstacle avoidance path planning of sudden static obstacle UAV.Firstly,the flight path planning model of sudden static obstacle was established by adding sudden static obstacle during the flight of UAV according to the global flight path.Considering the real-time requirement of local flight path planning,variable step size strategy and gravitational field are introduced into Rapidly exploring Random Tree(RRT)algorithm to improve search accuracy and convergence speed.Greedy algorithm and fitting interpolation principle are used to smooth the local track planned by RRT algorithm.Finally,the improved RRT algorithm is used for local flight path planning of burst static obstacles and compared with the classical RRT algorithm.The test results show that compared with the classical RRT algorithm,the improved RRT algorithm has better real-time performance and the planned flight path is more in line with the flight requirements of UAV.(3)Local obstacle avoidance path planning of sudden dynamic obstacle UAV.Based on the research of local obstacle avoidance path planning of sudden static obstacle UAV,dynamic obstacle is added into the global trajectory environment model,and the collision model between UAV and dynamic obstacle is established.The improved RRT algorithm is applied to local path planning,and the experimental results show that this method can effectively solve the problem of local obstacle avoidance path planning for sudden dynamic obstacle UAV.