Research On Power Inspection Technology Of Unmanned System Based On Visual SLAM

The Unmanned systems are increasingly used in inspection solutions for power infrastructure.They are lightweight,flexible,and autonomous and can achieve better power infrastructure inspection based on reducing labor costs.With the development of social economy and science and technology,and the rapid development of the power grid,transmission lines are gradually standardized and integrated.The current distribution of transmission lines is mainly divided into overhead transmission lines and underground cables.Patrol inspection has become an essential means of investigating hidden dangers in underground cable tunnels.For overhead transmission lines,transmission poles and towers,as a critical component of the overhead transmission lines,need to withstand external forces and conductor stress.Its stable operation directly affects the safety and reliability of the entire power supply network.If the poles are tilted and fail to meet the safety standards,the process of the transmission lines will cause safety hazards and power shortages.In summary,prevention is better than before,timely inspections of poles and cables to ensure quality is of great significance to the safe generation of electricity and people’s lives.This article is mainly aimed at the scene where the power transmission tower is inclined,equipped with unmanned systems(especially UVA),based on the improved ORB-SLAM2 algorithm to realize the routine inspection tasks of unmanned systems.At the same time,it can recognize its positioning based on the surrounding environment information.Furthermore,transmit the surrounding environment information together with its location information to the staff operating station to achieve timely warning.The main research content of this article includes the following three aspects.First,a basic introduction to the visual SLAM algorithm.According to the classification of sensors,SLAM algorithms can be roughly divided into visual SLAM led by optical sensors and laser SLAM led by laser sensors.In the current research and production activities,the visual SLAM algorithm is widely used.This article focuses on the visual SLAM algorithm,including the basic principles of the algorithm,the main framework of the algorithm,and the functions of each part of the framework.In addition,the optimization method is emphasized,and the difference in results due to different sensors is analyzed.Then,the research and application status of the inclination measurement of power transmission towers are introduced.First,it summarizes the key points and deficiencies of the past measurement methods and the current mainstream measurement schemes.Next,propose an inspection plan that considers safe areas.Finally,the advantages of ORB feature matching are studied,and an optimization algorithm of RANSAC,a mismatch removal algorithm based on traditional ORB feature matching,is proposed.Finally,the experimental verification of the aforementioned theoretical part is carried out.The ORB feature is compared with the SURF and SIFT feature matching by exploratory analysis and comparison,and the results are obtained and analyzed.Using ORB feature matching and progressive consensus algorithm and ORB-SLAM2 optimization framework,according to the custom monocular data set,using ROS as the experimental platform,the transmission tower inspection simulation experiment was carried out to verify the feasibility of the scheme.