Research On Route Planning Of Rotor UAV For Power Inspection

With the continuous expansion of the scale of China's power grid,the inspection and maintenance work of the power grid is facing more severe challenges.Traditional manual inspection methods are inefficient and inspectors have high work intensity and high danger.As an emerging inspection method,the flexibility and inspection efficiency of UAV(Unmanned Aerial Vehicle)inspection have incomparable advantages over manual inspection.The current UAV inspection mainly relies on remote control by ground operators,which needs high technical requirements for UAV operators.Automatic UAV inspection can further improve inspection efficiency and cost,so the realization of automatic inspection of UAV is ra current reseach hotspot.Autonomous route planning for UAV is one of the core technologies for automatic inspection of UAV.Studying the route planning methods of UAV is of great significance for achieving automatic inspection of UAV,improving inspection efficiency and reducing inspection cost.This thesis uses the power inspection as the task background to study the route planning methods of the UAV.The hierarchical planning structure which contains global planning and local planning is adopted to realize the autonomous route planning of the UAV under the three-dimensional digital elevation model.The main content of the thesis is as follows:(1)The research background and research significance of UAV track planning for electric power inspection is introduced,as well as the domestic and foreign research status of UAV inspection technology,UAV power inspection,and some common track planning methods.The task requirements of UAV power inspection and the problem of inspection path planning are analyzed,and the content of planning task is determined as the inspection of power equipment that is difficult for people to reach in complex terrain.Besides,simplified drone kinematics model is established,and the constraints of UAV performance are analyzed and determined.After that,three-dimensional elevation environment model that simulates the mountain terrain is established to provide a basis for subsequent route planning.The overall steps of route planning are determined according to the thought of hierarchical planning.(2)According to the overall steps of route planning,particle swarm optimization algorithm is selected for the global route planning of UAV under known environmental information.The characteristics of particle swarm algorithm and bacterial foraging algorithm is analyzed.Aiming at the defects of particle swarm optimization algorithm,the chemotaxis operator and migration operator of bacterial foraging algorithm is introduced into the iterative process of particle swarm algorithm.The improved algorithm is applied into the UAV route planning problem,particle swarm optimization three-dimensional path planning model is established and the fitness function is obtained by combining the UAV performanceconstraints.The route planning simulation experiments shows that the improved algorithm solves the problem that the particle swarm algorithm is easy to fall into a local optimum,and the stability and efficiency of the algorithm are significantly improved,and a better route can be planned.Finally,based on the planning results of the improved particle swarm algorithm,the B-spline curve method is used to smooth the route.(3)The local route replanning in the case of sudden obstacles when the UAV is flying along a predetermined route is studied.The shortcomings of the traditional artificial potential field method is analyzed,and the influence factor of the distance between UAV and the target point and the relative speed factor of UAV and the obstacle are introduced into the traditional repulsive field function to solve the target unreachable problem and enhance the dynamic avoidance of the potential field method.Besides,the line potential field force is introduced,so that the UAV can smoothly return to the predetermined route after avoiding obstacles.Simulation experiments in two-dimensional and three-dimensional environments are carried out,and the comparative experimental result shows the feasibility and effectiveness of the algorithm.