Pose Estimation Of Space Noncooperative Target Based On Lidar

As the demand for space and aerospace activities grow,countries have the degree of importance measurement and motion estimation of non-cooperative targets in the earth’s low track,especially those in the low-track non-cooperative targets.The number of satellites in various countries will climb year by year,so that the demand for the capture of high-value spacecraft,the requirements of space fragment cleaning,and the non-cooperative target racks,and the non-cooperative target.Estimate the research of related neighborhoods evolved into a hot field.The application prospects in this area are very broad,so the extensive sensor of the laser radar is used to study the relevant problems of space non-cooperative target positioning estimation.First,the API for receiving data from lidar is designed and PCL is added to the SDK.The data structures and algorithms in PCL are used to process the point cloud data.Filters based on features of the point from lidar and library functions in PCL can be carried out to suppress noise in the process of acquiring lidar data,at the same time the features of non-cooperative target are retained.Discrete frame pose estimation for non-cooperative target is executed using the traditional ICP algorithm,the ICP algorithm based on Kd-tree and the ICP algorithm combined with SAC-IA coarse registration.The performance of the three algorithms is comparedconsecutive frames pose estimation is performed by using the continuously obtained non-cooperative target point cloud frames.Pose estimation for noncooperative target based on lidar is divided into two parts: initial frame pose estimation algorithm and continuous frame pose estimation algorithm.The results of initial frame pose estimation are used as initial pose of continuous frame pose estimation algorithm.The point cloud set at the initial time is used as the initial source point cloud set for registration to obtain the result of pose estimation of the continuous motion between non-cooperative target and lidar.Finally,non-cooperative target pose monitoring platform is designed.Realtime display of non-cooperative target pose data on different devices,real-time curve drawing and result of pose estimation preserving are designed as the main functions.Network connection is established between the program of pose estimation algorithm and hose computer software within the same LAN.Visualizing the result of position estimation makes the relative motion of noncooperative target more intuitive than that of lidar,which provides technical support for future applications in the field of non-cooperative target position estimation.