Two-dimensional Lidar Scanning System Design And SLAM Research

Lidar,as an active scanning detection technology,can quickly obtain information about the surrounding environment.With the widespread application of lidar in robotics,unmanned driving and other fields,lidar markets at home and abroad are developing rapidly.In view of the high price of mature lidar products on the market,the monopoly of foreign technology and the lack of open source control algorithms,this topic has carried out the design and research work of lidar.In this paper,based on the indoor environment perception of mobile robots,combined with the on-board environment and actual application requirements,a laser ranging module developed by the research group is used to design a two-dimensional lidar scanning system.The designed two-dimensional lidar has close performance compared with the same type of products on the market,and the control algorithm is open source,but the cost is reduced by about half,which has good practical value.The main contents of the thesis are as follows:(1)Designed and built a two-dimensional lidar scanning system.The overall design scheme of the two-dimensional laser radar scanning system is given,the hardware of each module of the scanning system is selected and designed,and the design of the laser radar integration mechanism is completed;how to establish a CAN bus between the STM32F103ZET6 control chip and the scanning motor For communication issues,the CANopen network composed of the STM32F103ZET6 control chip and the scanning motor was configured,and the control chip to the scanning motor was realized under the CANopen protocol framework.(2)Completed the two-dimensional lidar scanning test experiment and designed the upper computer software of the lidar.Compile the control and data processing program of the two-dimensional lidar prototype to realize the functions of point cloud data data reading,coordinate transformation,de-noising filtering,etc.On this basis,two scenes are scanned and imaging experiments are performed,and real experiments are obtained.The three-dimensional point cloud data of the scene was analyzed and compared before and after the data processing;Under the Visual Studio 2017 platform,a two-dimensional lidar host computer was designed based on the C# language,and the host computer scanning test experiment and error analysis were completed.(3)Model the mobile robot,build its odometer motion model and lidar observation model under the mobile robot system,and select the grid map as the map representation model for this topic,and theoretically derive the update process of the grid map,and The Gmapping open source algorithm commonly used in indoor mobile robot SLAM(Simultaneous Localization and Mapping)is theoretically deduced.(4)The Gazebo simulation environment is constructed based on the ROS system,and the Gmapping algorithm is used for simulation experiments of mobile robot map construction;the Dash Go D1 mobile robot real environment test platform is built,and the mobile robot laser SLAM construction environment map experiment is performed,and the robot's mapping error is analyzed.The experimental results show that it is used the robot mapping accuracy of the two-dimensional lidar meets the design requirements,fully verifying the engineering practicability of the two-dimensional lidar.