ROS学习笔记之——3D map merge

Posted 2021-12-12 gwpscut

之前博客《ROS实验笔记之——基于cartographer的多机器人SLAM地图融合》已经实现了2D map merge，也做了大量这方面的demo

two

基于两个turtlebot2的多机器人SLAM地图融合

本博文看看3d map merge方面的资料

首先呢有CCM-SLAM（https://github.com/VIS4ROB-lab/ccm_slam）这个在之前的博客《学习笔记之——激光雷达SLAM（LOAM系列的复现与学习）》也介绍过了，以及做过复现。同时该组开开源了covins（https://github.com/VIS4ROB-lab/covins）。感觉这两者的差异估计不是很大。当然其框架有很多是值得借鉴的。

除了上面的以外，ros本身就有一个开源的包map_merge_3d （map_merge_3d - ROS Wiki）

其框架如下图所示

订阅的topic为<robot_namespace>/map (sensor_msgs/PointCloud2)

而发布的则是map (sensor_msgs/PointCloud2)

而offline融合的使用的方法是（比如采用SHOT描述子）融合三个map

rosrun map_merge_3d map_merge_tool --descriptor_type SHOT map1.pcd map2.pcd map3.pcd

Transformations between maps are estimated by feature-matching algorithm and therefore it is required to have sufficient amount of overlapping space between maps to make a high-probability match. If maps don't have enough overlapping space to make a solid match, merger might reject those matches.

参考资料

map_merge_3d - ROS Wiki