A-LeGO-LOAM

Introduction

Advanced implementation of LeGO-LOAM[1].

loop-closure enabled map cloud loop-closure enabled trajectory

Usage

Input

• Point Cloud(/lslidar_point_cloud)

Output

• (/laser_cloud_surround)
• ...

Run the package

1. Run the launch file:

roslaunch alego test2.launch

2. Play existing bag files test_0515.bag:

rosbag play test_0515.bag --clock --topics /lslidar_point_cloud

3. Save map:

rosservice call /save_map

ImageProjection

这 outlier 也太多了（将近 1/3），而且原始数据中有很多点相邻太近（将近一半），不知道是不是雷达的问题

LaserOdometry

robo_0529.bag 粗略测试，1 step 优化时长 7915ms(1.90ms/frame)(10 iterations)

2 step 优化时长 8888ms(2.13ms/frame)，效果更好(surf 5 iterations, corner 10 iterations)

不知道为啥，occluded points 就是比 lego 多，也是无语了，而且真要是标记为 occluded 的话 corner feature 就太少了，匹配起来效果很差。 无语了，原来是因为 cloud_msg 里的 segmentedCloudColInd 是 uint，进行算术运算再赋给 int 出了问题，然后 col_diff 就 gg 了。

LaserMapping

注意回环后 map2odom 要及时更新

Dependency

• gtsam(Georgia Tech Smoothing and Mapping library, 4.0.0-alpha2)
• ceres

TODOs

• parameterize hard coded parameters.
• find out nodelet crush problem.
• adjust motion distortion.

Referecen

1. https://github.com/RobustFieldAutonomyLab/LeGO-LOAM
2. https://github.com/HKUST-Aerial-Robotics/A-LOAM
3. Zhang J, Singh S. LOAM: Lidar Odometry and Mapping in Real-time[C]//Robotics: Science and Systems. 2014, 2: 9.
4. Zhang J, Singh S. Low-drift and real-time lidar odometry and mapping[J]. Autonomous Robots, 2017, 41(2): 401-416.