Autonomous Navigation And Inspection Strategy Of Rotorcraft UAV For PV Power Station

China has the world's largest photovoltaic(PV)installed capacity.The installed capacity is increased year by year,and the problem of PV power stations inspection is increasingly prominent.In recent years,quad-rotor unmanned aerial vehicles(UAVs)have developed rapidly,showing great potential in high-altitude search and monitoring.The efficiency and automation level of PV power stations inspection can be greatly improved by using UAV.Taking the autonomous navigation and inspection of UAV for PV power stations as the background,this paper studies and explores the path planning of PV power stations,infrared visual navigation of UAV,and wireless communication between UAV and ground stations in the process of autonomous navigation and inspection of UAV.According to the distribution characteristics of PV power stations,combined with the purpose and requirements of UAV inspection,the path planning problem of PV power station is transformed into the path planning problem between regions and the path planning problem within regions.The former is transformed into a multi-traveling salesman problem,which is solved using genetic algorithms,to obtain the task allocation and inspection order in each sub-region.The latter is transformed into an overlay path planning problem,which is solved by coordinate transformation,circumscribed rectangle construction,path point generation,and PNPoly algorithm,to get path points in the subregions.Experimental results verify the effectiveness of the algorithm.An autonomous navigation and inspection strategy based on infrared vision is proposed to make up for the shortcomings of navigation and inspection strategy based on path points.In order to obtain the position of PV strings,a PV strings recognition and location algorithm combining color features,gradient features and shape features is designed.In order to avoid missing or re-checking of strings during multi-row inspections,the reconstruction algorithm is designed.In order to realize the trajectory tracking and turning of the centerline of PV strings,a flight speed control scheme and turning scheme based on the position information of PV strings are proposed.The accuracy and effectiveness of the proposed algorithms are verified by experiments.A wireless communication system between UAVs and the ground station is established to monitor UAVs and increase the flexibility of inspections.This communication system includes four functional modules: file / message transmission,status information transmission,image transmission,and command control.The UAV client uses 4 independent nodes to implement different functional modules,and the ground station server uses a multi-threaded server model.At the same time,the implementation method of each sub-function module is proposed,and the interaction protocol between server and client is designed.Experiments show that the communication system is effective and stable during the UAV flight.