Detection And Reconstruction Of Nuclear Radiation Environment For Radioactive Source Location

Since the rapid development of nuclear technology,nuclear technology has been mastered by many large countries and applied in various fields.Due to the early start of nuclear energy technology,today’s nuclear decommissioning and nuclear accidents occur frequently,seriously endangering the safety of society and the public.However,due to the impact of ionizing radiation from nuclear sources,relevant staff cannot directly enter the radiation environment to complete corresponding work.Therefore,this paper designs a system based on Multi-sensor data fusion methods such as γ camera,3D laser radar and IMU(Industrial Measurement Unit)are used to build radiation environment map and calculate the distribution of radiation area in the environment,so that relevant staff can complete their work.The main work of this paper is as follows:(1)The multi-sensor environment information detection platform of lidar,IMU and gamma camera is designed,and the mathematical model of the sensor in the system is deduced.The checkerboard calibration board is used to jointly calibrate the γcamera and the lidar,and then the LIDAR-IMU joint calibration is performed using the pose estimation relationship between the IMU and the lidar.The pose transformation relationship between the sensors is solved through the calibration between the sensors,and the sensor data is fused in the ROS operating system.(2)Aiming at the influence of the radiation environment on the visual system,a tight coupling method of lidar and IMU data is designed to build the radiation environment map.Firstly,the feature point extraction between the lidar point cloud data and the frame-local map point cloud matching are used as the position and attitude estimation data of the lidar,and then the pre-integration of IMU data is used as the position and attitude estimation data of the inertial measurement unit.Finally,the two closely coupled sensor data and loop detection are used as factors to optimize the system posture and map construction.(3)The fusion method of radiation information and laser point cloud is designed.The data fusion relationship between gamma camera and laser radar is used to obtain the radiation point cloud in a single frame radar point cloud.Combined with the system position and attitude estimated by regional reconstruction,the distribution information of radiation point cloud and system sensor perspective in the map is obtained.The overall information of the radiation point cloud region is obtained by combining the radiation point cloud information from multiple perspectives.The radiation contour is used to limit the point cloud in the radiation area,and the method of minimum bounding box and view line is used to divide and locate the radiation source in space.(4)The experimental stage is divided into two parts: environment reconstruction and radiation source location.First of all,the radiation environment reconstruction experiment uses the same lidar and inertial measurement unit data for reconstruction,and compares the effects of the LOAM and LIOM algorithms,reflecting the robustness of the environment reconstruction method in this paper.Finally,the space division and location of the radiation region are compared with the actual source coordinates,which proves the effectiveness of the method in this paper.