Integration And Quality Control Of Mobile Laser Scanning Mapping System

Geographic information technology,communication technology,perception technology and super-computing technology are the key technologies in the construction of smart city.Among them,geographic information technology is the key to gather data and information.High-speed and high-precision acquisition of urban spatial geographic information data for the construction of smart city is a hot issue of current research.Mobile laser scanning mapping system based on multi-sensor integration is a main technology of urban spatial geographic information acquisition.It is a new,comprehensive and highly dynamic mapping method,and has been widely used in street scene data acquisition,municipal management and road survey.However,there are still many problems in the popularization and practical application of the mobile laser scanning mapping system,which include:(1)the complete mobile mapping system is expensive and the cost of use is relatively high.(2)The position relationship between multi-line lidar and inertial navigation system is biased,resulting in the migration of point cloud data.(3)Under the complex environment,the quality of system point cloud data is seriously degraded.Therefore,it is of great practical significance for the popularization and application of mobile laser scanning mapping system to reduce the cost of mobile mapping system,study the calibration of multi-line lidar external parameters and the quality control method of point cloud data accuracy in complex environment.Based on this,the main work accomplished in this paper is as follows:1.A detailed research on development status of mobile mapping system at home and abroad.Described research present summary on external parameters and quality improvement of point cloud of mobile mapping system.The related theoretical foundation of mobile mapping system is introduced.Then,the sensor is connected and integrated with the system design.2.Completed the design integration and implementation of mobile laser scanning mapping system.The overall design of the system is completed by modularized decomposition of the system,and the sensor selection strategy is introduced,using dual antenna attitude determination technology to assist IMU initial alignment to improve IMU measurement accuracy.A time synchronization method based on second pulse triggering is proposed for the unification of time datum in the system and introduceing the interpolation algorithm.Aiming at the unification of spatial datum,the process of coordinate transformation is deduced,and the initial value calibration of external parameters based on direct measurement method is introduced.For the special data format of multi-line lidar,based on the analysis of data analysis method,the data analysis program is compiled on the platform of MATLAB.Finally,the integrated mobile laser scanning mapping system designed in this paper can realize the function of data acquisition and data fusion.3.Study on a one stop fast calibration method for multi-line LiDAR external parameters based on rectangle marks.Based on the analysis of the scanning characteristics of multi-line lidar and the limitations of traditional artificial markers,the paper proposes to use standard rectangular cardboard as an artificial marker,and the corner as a common point.The corner extraction,a method of fitting the edge of the mark in the point cloud by RANSAC and taking the edge intersection as the corner is proposed,while there is no intersection point,the middle point at the nearest edge distance is the corner point.To slove the external parameters Kabsch algorithm are used to solve the optimal selection matrix and translation parameters,and compared with the direct measurement method.Finally,the external parameters are brought into the data fusion process of road survey experiment to verify the accuracy of placement parameters.4.Research on modification method of point cloud data quality in mobile mapping system under complex environment.Aiming at the quality control of single frame laser scanning data,a point cloud filter based on measurement range constraint is proposed to remove redundant data from the original point cloud,and a method to improve the ranging error of lidar considering the correction of point cloud parameters is proposed.To improve the accuracy of point cloud,a data-driven quality control method based on known constraints is proposed.The corresponding relationship between the known control points measured by total station and the control points in the point cloud is established.A coordinate transformation model is constructed to solve the conversion parameters.The data of point cloud is directly modified by this parameter,and the performance of chemical effect of the number and distribution of control points are analyzed.For the quality control of fused point cloud data,a discrete point method based on improved k-nearest neighbor algorithm and an external solitary point elimination method based on covariance matrix are proposed.Finally,the effectiveness of the above methods for quality control is verified by experiments.