Re: Combine 3D Lidar and RGB-D Camera to create a single 3D point cloud (sensor fusion)

Posted by Alex Frantz on
URL: http://official-rtab-map-forum.206.s1.nabble.com/Combine-3D-Lidar-and-RGB-D-Camera-to-create-a-single-3D-point-cloud-sensor-fusion-tp10162p10347.html

Hi again,

Sorry for the late reply on this thread. Thanks a lot for taking the time to reply to my questions!

I have tried to incorporate your proposed solution and here are my findings:

1. The camera's intrinsic parameters were indeed wrong. I too cannot use any other values than the one you suggested. Unsure as to why this is the case currently... The camera's depth cloud is not properly aligned with the floor.

2. Regarding the launch config for rtabmap to produce a point cloud which merges inputs from both the RGBD camera and the 3D LiDAR, I have run into some issues. When I use the code you provided, the point cloud generated only seems to use the LiDAR data and ignores RGBD input. However, I noticed that you had set the parameter:

RGBD/CreateOccupancyGrid : False.

I experimented and set this one to true, and at the same time added another parameter: Grid/Sensor : 2 , which to my understanding, allows RTABMAP to generate a point cloud using both 3D LiDAR and Depth data. With these two changes, I have been able to get what I need. Yet, I am not sure if this is the correct approach or if I missed something in your response to get the results I desire.

Would be possible to confirm or correct my approach please?

Cheers,

Alex