Intelligent unmanned aerial vehicle inspection system and method of tunnel

Abstract

The invention discloses an intelligent unmanned aerial vehicle inspection system and method of a tunnel. On the basis of combination of the unmanned aerial vehicle technology and the image processing technology, the unmanned aerial vehicle is applied to tunnel inspection work. According to the invention, various sensors are added to an unmanned aerial vehicle platform; with data collected by a monocular black-and-white camera and an inertial measurement unit (IMU), precise positioning and navigation of the unmanned aerial vehicle in the tunnel are realized by means of multi-sensor data fusion according to a VINS-Mono positioning mapping algorithm; a 2D laser radar collects point cloud data to realize estimation of a tunnel wall direction and UAV obstacle avoidance; a high-resolution camera collects a tunnel image, a motion trajectory of the camera is restored based on an SfM algorithm, and a three-dimensional model of the tunnel is generated by combining a high-definition image, thereby realizing the detailed tunnel inspection task. Compared with the traditional tunnel inspection way relying on the personnel and vehicles, the intelligent unmanned aerial vehicle inspection system and method have advantages of high automation degree and high work efficiency and can be widely applied to tunnel inspection work.

Description

technical field [0001] The invention belongs to the technical field of image processing and unmanned aerial vehicles, and relates to an unmanned aerial vehicle inspection system and method, in particular to autonomous positioning, surrounding environment perception, unmanned aerial vehicle path planning and obstacle avoidance of an unmanned aerial vehicle in a tunnel environment , Image three-dimensional reconstruction system and method. Background technique [0002] In recent years, with the continuous development of our country, the domestic economic level has grown rapidly every year and the comprehensive national strength has been continuously improved, and the construction capabilities of various infrastructures have been greatly improved. With the completion of a series of large-scale tunnel projects such as the immersed tube tunnel of the Hong Kong-Zhuhai-Macao Bridge, the Yangtze River Tunnel on Sanyang Road in Wuhan, and the Guanjiao Tunnel on the Qinghai-Tibet Rail...

AI Technical Summary

Problems solved by technology

[0004] However, the environment inside the tunnel is extremely complex and harsh, which poses great challenges to drone operations

Due to occlusion inside the tunnel, the GPS signal cannot penetrate. Common commercial UAV systems that rely on GPS signals for UAV pose estimation cannot fly in the tunnel.

In addition, factors such as the electrical system of the subway tunnel and the magnetic layer near the tunnel will cause the magnetic field in different sections of the tunnel to change drastically, and the drone cannot calibrate the direction

In addition, the spatial structure of the tunnel is relatively narrow compared to common drone flight scenarios, such as outdoors. When the drone is flying, it is easily disturbed by the airflow rebounded from the wall, which seriously interferes with the stability of the drone, and may lead to a crash in severe cases. machine

Images