Research On Intelligent Obstacle Avoidance Path Planning Of UAV And Its Application In Autonomous Power Line Patrol

With the increases of transmission voltage level and the increase of transmission line length in China,more and more attention has been paid to the implementation of power line patrol tasks by UAV.At present,the power line patrol work is mainly carried out by the UAV manually controlled by the staff.If the UAV can fly autonomously along the power line and carry out inspection tasks,the line patrol efficiency of UAV will be improved exponentially.In the autonomous flight system of UAV,the autonomous obstacle avoidance path planning of UAV is one of the main core technologies.Therefore,the research on the method of autonomous obstacle avoidance path planning of UAV can not only reduce the manpower,material and financial resources of patrol line,but also has a very important significance to improve the efficiency of UAV patrol line.When an UAV is performing patrol tasks,its obstacle avoidance path planning involves the processing of many different information,which requires the performance of the UAV’s computer especially high.Therefore,this paper divides the autonomous patrol process of UAV into two layers according to the characteristics of the UAV patrol power line:the first layer is the global obstacle avoidance path plan before the UAV patrol line,which finds a better obstacle avoidance path mainly based on the information of the global environment;the second layer is the local obstacle avoidance path plan when the UAV patrols along the global obstacle avoidance path,which only considers the instantaneous environmental information.In order to avoid unexpected threats in time and plan obstacle avoidance tracks online,the difficulty of obstacle avoidance track planning is reduced due to the UAV ’s own factors.The main contents of this paper are as follows:(1)This paper analyzes the development status of UAV line inspection technology at home and abroad,and summarizes the research status of common obstacle detection technology and obstacle avoidance path planning algorithm of UAV.(2)By analyzing the environmental constraints of the patrol task area and classifying the patrol inspection according to the route,the work content of regular patrol is selected as the content of UAV patrol in this paper.Select the appropriate patrol model based on the unmanned aerial vehicle flight environment and its own performance constraints,and the representation of the unmanned obstacle avoidance track is selected.By studying the current popular obstacle detection technology,this paper determines the use of binocular vision technology to detect unexpected threats encountered in local obstacle avoidance,and chooses the geometric description method to build the obstacle model that can be recognized by the computer.According to the characteristics of the actual power line environment,the function simulation method is selected to establish the environment model of UAV patrol line.(3)According to the characteristics of global obstacle avoidance path planning,an improved target point selection strategy RRT(P-RRT)algorithm is proposed as global path planning algorithm for UAV.For the characteristics of local obstacle avoidance path planning,an RRT algorithm based on pre-planned path optimization is proposed to avoid local static sudden threats and an improved artificial potential field method to avoid local dynamic threats.The simulation experiments are carried out in the same environment,and the advantages of the improved algorithm are proved by comparing the experimental results.(4)According to the hierarchical route planning method proposed in this paper,an experimental simulation is carried out in the established patrol task area environment model to verify the effectiveness of the proposed method in power autonomous patrol.