Re: ros-noetic usage

Posted by matlabbe on
URL: http://official-rtab-map-forum.206.s1.nabble.com/ros-noetic-usage-tp9609p9640.html

I would suggest to make your own launch file instead of modifying rtabmap.launch because it takes too long for reviewers to spot the differences.

The screenshots of previous post have been re-scaled, make it impossible to read the text.

I have difficulty to understand where is the problem coming from, is it rtabmap's vo? the actual fusion? I would try to compare with and without odometry fusion first, as right now I cannot say if it is coming from rtabmap's vo or from fusion.

Looks like the RGB/Depth frames are wrongly synchronized, I'll suggest to follow suggestions in the warning.

If you are interested on the maximum VO accuracy with D435, use stereo mode with left/right IR images (with IR emitter disabled).

cheers,
Mathieu