Research On The Structural Positioning Consistency Method Of Flying Robot For Transmission Tower

With the rapid development of State Grid construction,the traditional manual patrol is far from matching the requirement of the daily maintenance and overhaul of transmission lines and towers.The usage of unmanned aerial vehicles(UAVs)based inspection method can improve the efficiency and reduce the maintenance costs,which has important research significance and wide application background.Precise localization technology is the basis for UAVs to realize various inspection functions.This work studies on UAVs’ localization problems in the inspection process of transmission lines and towers,and mainly focuses on the environmental assessment-based technology for laser scanning matching and IMU fusing localization.And mainly disscusses some key points which include parametric reconstruction of transmission line,parametric method of secant slope characteristic of transmission line,construction of environmental evaluation function and dynamic update strategy of pose.Firstly,this work studies the outlier point removal method in laser scanning data,and reduces the number of point clouds by VoxelGrid method which could improve the computational efficiency of the matching algorithm.Secondly,An inertial measurement unit(IMU)and correlative scan matching(CSM)fusing localization method was presented based on environmental assessment.the UAV’s initial pose was estimated by the IMU,the environmental evaluation function was constructed based on the current scan.the output of the function was compared with the environmental difference threshold to update the confidence of CSM localization,and the UAV’s pose update strategy was dynamically selected with the confidence,which can improve the matching accuracy of CSM in case of small environmental difference.Finally,a secant slope characteristics-based method for transmission line was presented to reduce the integral accumulated error of IMU in case of highly environmental similarity,while the data preprocessing method was used to extract the transmission line secant,and the sceant slope feature is matching with the parameterized reconstructed transmission line map to estimated the UAV s translation vector.Experimental results show that the method is more adaptable,accurate and efficient than the traditional method.