lidar_undistortion_2d
lidar_undistortion_2d copied to clipboard
Published
20 hours ago •
LiuLimingCode
LiuLimingCode
lidar_undistortion_2d
a ros package for lidar motion compensation
Introduction
读取odom数据对2D激光雷达数据进行运动畸变校正。
This ros package uses odom transform data to correct motion distortion of a 2D LIDAR in real time。
Result
在图片中，黄色方框代表机器人的位姿，红色点云代表原始的激光雷达数据，白色方框代表经过运动补偿后的激光雷达数据。
in this picture, the yellow rectangle represents the pose of robot, the red poindcloud represents the origin lidar data, and the white pointcloud represents the lidar data after compensation.
Parameters in launch file
| 名称 | 类型 | 注释 |
|---|---|---|
| scan_sub_topic | string | 订阅的激光数据话题名 |
| scan_pub_topic | string | 经过运动畸变矫正后发布的激光数据话题名 |
| enable_pub_pointcloud | bool | 是否将校正后的数据重新封装为LaserScan消息发布 |
| pointcloud_pub_topic | bool | 经过运动畸变矫正重新封装LaserScan消息话题名 |
| lidar_frame | string | 激光雷达数据的坐标系 |
| odom_frame | string | Odometry数据的坐标系 |
| lidar_scan_time_gain | double | 激光雷达单次扫描时间系数（正常情况下是1.0，但是有些激光雷达的驱动包在计算scan_time时有问题，所以这里乘一个系数） |
Test with rosbag
- compile the project and
source devel/setup.sh - execute the following command
roslaunch lidar_undistortion_2d test_lidar_undistortion_2d.launch enable_undistortion:=true
- find
/bag/sensor_data.bag
rosbag play --clock --pause sensor_data.bag
remind: '--pause' is essential. otherwise it may lead to error.
- result
the gif showed below represents location with orign lidar data.
the gif showed below represents location with undistortion lidar data.
Reference
https://github.com/elewu/2d_lidar_undistortion
深蓝学院SLAM教程
LiuLimingCode