Research On Simultaneous Localization And Mapping Of Smart Car Based On Multiline Lidar

In recent years,with the rapid development of auto-driving technology,positioning and environmental awareness as the key technology of auto-driving have received much attention.Among them,the simultaneous localization and mapping(SLAM)technology enables the vehicle to achieve its own positioning and perception of the environment independently,which is an important basis for the automobile to achieve full autonomy.Based on multiline lidar,the simultaneous localization and mapping of smart cars is studied.Firstly,three key technologies including feature extraction,point cloud registration and non-linear optimization are introduced,and their implementation methods are analyzed.Next,build a platform based on the robot operating system,select the working frequency of lidar according to the parameters and working characteristics of lidar,and complete the preprocessing of point cloud data.Radius filter and pass filter are used to remove outliers,reduce point cloud noise,and voxel filter is used to down-sample the original point cloud data to reduce the amount of data.Finally,the plan of SLAM of smart cars is completed.An Inertial Measurement Unit(IMU)pre-integration model is established to eliminate point cloud motion distortion by using IMU pre-integration.The feature-based point cloud registration method is used for inter-frame registration,and the LM algorithm is used to obtain the initial position estimation.Fuse IMU with lidar to obtain optimal position estimation using the result of interframe registration as initial value.To solve the problem of large accumulative error in continuous inter-frame registration,the current frame point cloud is registered with the local map in a low frequency way to reduce the accumulative error.And based on the registration results,the current frame point cloud data is updated to the point cloud map,and the environment map is built.The registration results are fused with the optimized position estimation to obtain the accurate position of each frame point cloud in the global coordinate system,thereby updating the vehicle location.In order to verify the feasibility of this scheme,an smart car experimental platform is set up for experimental verification.Using the positioning results of the vehicle integrated navigation system and satellite images,the positioning and environment map construction results of this scheme are analyzed.The experimental results show that this scheme is feasible.