基于惯性测量单元的激光雷达点云融合方法
|
摘要
针对16线激光雷达环境感知过程中,点云数据稀疏,导致对目标检测和识别困难的问题,提出了一种基于惯性测量单元(InertialMeasurementUnit,IMU)的激光雷达点云融合方法。建立了激光点云数据的融合模型,有效利用历史点云数据与历史检测结果,获得较多的环境信息,提高了目标物的检测精度。利用16线激光雷达与自研的IMU传感器进行实验验证,结果表明能够实现激光雷达点云的融合,进一步提高激光雷达对目标物的检测能力,并且以较低的硬件成本,实现更加高级的环境感知能力,对无人驾驶等技术的研究具有实际应用价值。
Aiming at the problem that in the process of using 16-line laser radar to realize environment perception, the point cloud data is sparse, which leads to the difficulty of target detection and tracking, a new method of LiDAR point cloud fusion based on inertial measurement unit(IMU) is proposed. The method establishes a multi-frame LiDAR point cloud data fusion model, which can effectively use historical point cloud data and detection results to obtain more environmental information, and improve the detection accuracy and tracking ability of target objects. 16-line laser radar and the self-developed IMU sensor are used to conduct the tests. The results demonstrate that the proposed method can achieve the multi-frame fusion of the laser radar point cloud, and the detection and tracking ability of the laser radar can be further improved. And more advanced environment awareness is achieved with lower hardware costs, which shows that the method has practical application value for the study of driverless technology.
Aiming at the problem that in the process of using 16-line laser radar to realize environment perception, the point cloud data is sparse, which leads to the difficulty of target detection and tracking, a new method of LiDAR point cloud fusion based on inertial measurement unit(IMU) is proposed. The method establishes a multi-frame LiDAR point cloud data fusion model, which can effectively use historical point cloud data and detection results to obtain more environmental information, and improve the detection accuracy and tracking ability of target objects. 16-line laser radar and the self-developed IMU sensor are used to conduct the tests. The results demonstrate that the proposed method can achieve the multi-frame fusion of the laser radar point cloud, and the detection and tracking ability of the laser radar can be further improved. And more advanced environment awareness is achieved with lower hardware costs, which shows that the method has practical application value for the study of driverless technology.
引文
[1]赵一鸣,李艳华,商雅楠,等.激光雷达的应用及发展趋势[J].遥测遥控,2014,35(5):4-22.Zhao Yiming,Li Yanhua,Shang Ya’nan,et al.Application and Development Direction of Lidar[J].Journal of Telemetry,Tracking and Command,2014,35(5):4-22.
[2]林辉.基于车载多激光雷达的地图构建与障碍物检测[D].杭州:浙江大学,2017.Lin Hui.Map Building and Obstacle Detection based on Vehicle-mounted Multiple Lidars[D].Hangzhou:Zhejiang University,2017.
[3]刘家银,唐振民,王安东,等.基于多激光雷达与组合特征的非结构化环境负障碍物检测[J].机器人,2017,39(5):638-651.Liu Jiayin,Tang Zhenmin,Wang Andong,et al.Negative Obstacle Detection in Unstructured Environment Based on Multiple Li DARs and Compositional Features[J].ROBOT,2017,39(5):638-651.
[4]Shang E K,An X J,Wu T,et al.LiDAR based negative obstacle detection for fieldautonomous land vehicles[J].Journal of FieldRobotics(S1556-4959),2016,33(5):591-617.
[5]Morton R D.Positive and negative obstacle detection using the HLD classifier[C]//Ieee/rsj International Conference on Intelligent Robots and Systems.IEEE,(S8755-3996),2011:1579-1584.
[6]Larson J,Trivedi M.Lidar based off-road negative obstacle detection and analysis[C]//International IEEEConference on Intelligent Transportation Systems.IEEE,(S8755-3996),2011:192-197.
[7]蒋钰,谌海云,岑汝平.基于四元数的四旋翼飞行器姿态解算算法[J].制造业自动化,2015,37(23):77-80.Jiang Yu,Zhan Haiyun,Cen Ruping.Attitude solution algorithm for four rotor aircraft based on four element number[J].Manufacturing Automation,2015,37(23):77-80.
[8]Serban R,Wang S K,Lo M W,et al.Halo orbit mission correction maneuvers using optimal control[J].Automatica(S0254-4156),2002,38(4):571-583.
[2]林辉.基于车载多激光雷达的地图构建与障碍物检测[D].杭州:浙江大学,2017.Lin Hui.Map Building and Obstacle Detection based on Vehicle-mounted Multiple Lidars[D].Hangzhou:Zhejiang University,2017.
[3]刘家银,唐振民,王安东,等.基于多激光雷达与组合特征的非结构化环境负障碍物检测[J].机器人,2017,39(5):638-651.Liu Jiayin,Tang Zhenmin,Wang Andong,et al.Negative Obstacle Detection in Unstructured Environment Based on Multiple Li DARs and Compositional Features[J].ROBOT,2017,39(5):638-651.
[4]Shang E K,An X J,Wu T,et al.LiDAR based negative obstacle detection for fieldautonomous land vehicles[J].Journal of FieldRobotics(S1556-4959),2016,33(5):591-617.
[5]Morton R D.Positive and negative obstacle detection using the HLD classifier[C]//Ieee/rsj International Conference on Intelligent Robots and Systems.IEEE,(S8755-3996),2011:1579-1584.
[6]Larson J,Trivedi M.Lidar based off-road negative obstacle detection and analysis[C]//International IEEEConference on Intelligent Transportation Systems.IEEE,(S8755-3996),2011:192-197.
[7]蒋钰,谌海云,岑汝平.基于四元数的四旋翼飞行器姿态解算算法[J].制造业自动化,2015,37(23):77-80.Jiang Yu,Zhan Haiyun,Cen Ruping.Attitude solution algorithm for four rotor aircraft based on four element number[J].Manufacturing Automation,2015,37(23):77-80.
[8]Serban R,Wang S K,Lo M W,et al.Halo orbit mission correction maneuvers using optimal control[J].Automatica(S0254-4156),2002,38(4):571-583.