ROS-based Thermal Imaging Inspection Robot Design

It is very important to ensure the safety of the eletric transmission line for the production safety of the factory.In order to find out the hidden danger of fire,the transmission line is usually inspected manually and regularly.This method has some problems,such as low inspection efficiency,heavy workload,difficult detection,difficult data collection and so on.Inspection robot is the best choice to solve these problems.It can complete intelligent inspection,with high inspection efficiency,accurate detection and timely detection of problems,which has incomparable advantages over manual inspection.In this thesis,based on the inspection requirements of the underground pipeline equipment of a factory in Qingdao,a thermal imaging inspection robot based on ROS is researched and designed.The inspection robot uses the Lidar to obtain the surrounding environment information,builds the environment map and stores it in the industrial computer.During the inspection,the robot calls the environment map,and according to the location information of the set task point and starting point,as well as the static and dynamic obstacle information in the environment map,it plans the optimal operation path to carry out the inspection task.During the inspection,the infrared camera obtains the wiring temperature in real time and uploads it to the host computer.According to the detected temperature data,the infrared thermal image is generated and displayed in pseudo color to the host computer.The complete infrared image is displayed by image mosaic.The specific work of this thesis is as follows:(1)Based on the factory inspection requirements,a ROS-based inspection robot system is built.The overall structure and main functions of the system are designed,and the hardware devices supporting the system functions are selected.(2)The navigation system of inspection robot is established.The motion model of the robot is established and its motion process is analyzed.PID algorithm is used to control the speed of DC motor.The environment model suitable for robot SLAM is analyzed,the principle of RBPF-SLAM algorithm and the improved RBPF-SLAM algorithm are researched,so that the robot can build environment map.The principle of the Q-Learning path planning algorithm is researched,and the robot can plan the inspection path independently in the environment map.(3)The infrared thermal imaging system is completed.The principle of infrared thermal imaging is researched,and its mathematical model is analyzed.The reasons for the errors in the temperature measurement of the infrared sensor are researched,and the temperature output of the sensor is corrected by the calibration algorithm,and the grayscale image is generated.Through the image segmentation stretching algorithm,the grayscale image is optimized to enhance the contrast of infrared image.The functional relationship between grayscale and pseudo-color is researched,and the infrared thermal image is displayed by pseudo-color coding.The SURF image mosaic algorithm is researched,and the infrared image is mosaicked according to the feature similarity matching.The weighted average method is used to fuse the mosaic images.After mosaicking,the infrared image of temperature measurement scene can be displayed completelyFinally,the thermal imaging inspection robot is tested comprehensively.The test environment is built in the laboratory,and the key technologies such as map building,path planning,image processing and image mosaic are simulated to verify the effectiveness of the thermal imaging inspection robot scheme,which lays a solid foundation for the subsequent practical application.