Ask Your Question
0

Any suggestions for combining scans of different frequencies?

asked 2016-02-19 02:30:28 -0500

ruffsl gravatar image

I'm combining two scans, one extracted from a depth camera, and the other from a 2D LIDAR. But I'm getting some jitter from the depth camera's scan's contribution. I suppose it's to be expected when combining a 30Hz depth scan with a 40Hz LIDAR, but I was wondering if there was anything that could be done better. A qualitative example of the jitter artifact in RVIZ; red/LIDAR, green/depth-camera, yellow/combined. Note how combined lagges with the camera, and the LIDAR remains quite static. image description

The example you see here is a turtlebot simulation from gazebo, with a asus pointing forward, and a gpu_lidar pointing backwards, using this small PR of ira_laser_tools: #7. I made an issue #8 of this, but thought It could be general enough for a good old approach question here.

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted
0

answered 2016-02-19 06:27:03 -0500

This isn't trivial and especially with real sensors there will be additional problems due to noise (and different noise characteristic). That being said, I imagine using odometry (or z gyro/angular velocity data directly) during the merging process to rotate data appropriately is something that might help. For this to work correctly, the timestamps of the involved data have to be somewhat accurate, however.

edit flag offensive delete link more

Comments

Yep, I guess combining the scans with respect to the odom frame, then going back to the destination_frame might improve this. Perhaps an amendment around here?

ruffsl gravatar imageruffsl ( 2016-02-19 14:44:10 -0500 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools

2 followers

Stats

Asked: 2016-02-19 02:30:28 -0500

Seen: 145 times

Last updated: Feb 19 '16