i/rtabmap
1
0
Fork 0
mirror of https://github.com/introlab/rtabmap.git synced 2026-10-09 03:27:45 +08:00
Code Issues Packages Projects Releases Wiki Activity

Updated Change parameters (markdown)

matlabbe
2015-01-12 15:19:14 -05:00
parent 167608b68a
commit 5a4eb26678
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Change-parameters.md
+1 -1
@@ -94,7 +94,7 @@ I will show each panel from the Preferences dialog with the corresponding parame
[[doc/Tutorials/SpeedUp/RGB-D_SLAM.png]]
1. Must be checked for RGB-D SLAM.
2. We don't refine loop closure constraint transformations (computed visually) to save CPU time. While ICP 2D is useful on a robot with a 2D laser, 3D ICP suffers from difficult problems to detect (see [ICP page] for a discussion why ICP 3D is not used by default).
2. We don't refine loop closure constraint transformations (computed visually) to save CPU time. While ICP 2D is useful on a robot with a 2D laser, 3D ICP suffers from difficult problems to detect (see [ICP page](https://github.com/introlab/rtabmap/wiki/ICP) for a discussion why ICP 3D is not used by default).
3. When checked, the referential of the map is set according to the last node added to the map's graph, so the last map created.
4. I disabled local loop closure detection because I use the [local history approach](https://github.com/introlab/rtabmap/wiki/Change-parameters#odometry) for odometry. I've found that activating both at the same time results in a lower quality of the map. Local loop closure detection adds also more constraints in the graph, so more time required to optimize the graph (not ideal when RTAB-Map's update rate is fast).