Re: About Visual slam with lidar

Posted by matlabbe on
URL: http://official-rtab-map-forum.206.s1.nabble.com/About-Visual-slam-with-lidar-tp8798p8817.html

Hi,

First question yes, in that configuration.

For the second question, it is one or the other, you can set Grid/FromDepth to switch between lidar and camera. Note that using the lidar-only for occupancy grid, it is still possible to avoid 3D obstacles during navigation by feeding camera cloud to local costmap of move_base. Global plans are done in the clean lidar map, and navigation is using both, which is actually my favorite config.

Actually, in latest rtabmap version, Grid/FromDepth parameter is now called Grid/Sensor and camera-lidar config is supported. If
* Grid/Sensor=0, map is created from lidar,
* Grid/Sensor=1, map is created from depth image
* Grid/Sensor=2, map is created from both lidar and depth image

cheers,
Mathieu