Design Of Subway Tunnel Surface Defect Detection Robot

With the rapid development of urban rail transit,the safety condition detection of subway tunnels is becoming more and more important.In view of the existing subway tunnel inspection technology cannot obtain the tunnel crack depth information,difficult to distinguish between surface crack and stain marks such as the problem of background interference,this paper designs a car based on robot and structured light sensor crack identification technology has high precision,through the crack of the three-dimensional information,realize the reliable testing of subway tunnel crack.Firstly,aiming at the problem of vehicle-mounted detection of tunnel surface defect depth information,a hardware device for 3D information acquisition and data processing is designed based on the existing subway tunnel detection vehicles.On the basis of the existing positioning system of the detection vehicle,a new mechanical arm and structural light system are added.The software drive interface design of the mechanical arm and structural light sensor is completed,and the control and data acquisition of the mechanical arm and structural light sensor are realized.In order to solve the problem of obtaining 3D information of crack on tunnel surface,the trajectory generation algorithm of manipulator and the structural light acquisition algorithm are proposed.The trajectory generation algorithm of the robot arm is based on the information detected by the comprehensive detection vehicle to send and execute commands,generate the path of the robot arm,and then call the real-time Angle generation function of the robot arm to get the Angle information of the robot arm pose.Based on the path of the mechanical arm,the structural light acquisition algorithm collects the two-dimensional depth information of the crack on the tunnel surface during the end movement of the mechanical arm.To solve the problem of reliable identification of surface cracks in the tunnel,hand-eye calibration was carried out between the mechanical arm and the structural light sensor,and the Angle information generated by the mechanical arm and the 2D information output by the structural light sensor were combined for data processing.Finally,a 3D model was generated to obtain high-precision depth information of surface cracks in the tunnel.Finally,in order to verify the accuracy and effectiveness of the tunnel surface crack detection system,a laboratory model test is carried out.The experimental results show that the measuring range of crack on tunnel surface is up to 300 mm,the collection interval of crack length direction is up to 8mm,and the measurement error of crack depth is within ±2mm.Combined with the historical measurement data,the system can realize the trend detection of fine cracks on the tunnel surface.This paper proposes a subway tunnel surface defect new method of 3D information detection,can solve the existing crack defects detection based on machine vision technology,unable to distinguish between crack with traces of the water damage problem,to enhance the accuracy in surface of tunnel disease recognition,scientific guiding the maintenance of the tunnel,it is of great significance.