-
Notifications
You must be signed in to change notification settings - Fork 22
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
imaging sonar plugin frame issues #11
Comments
Thank you for the report! Could you reproduce these at a sonar tank set-up? |
I tried to do so using However, it doesn't publish /tf, which is required to visualize the sensor data in rviz, which is how I'm seeing that it's inconsistent. Do you have a test environment with the imaging sonar attached to a robot and a launch file running robot_state_publisher? |
I've been trying to debug this and make an example case of a robot with multibeam sonar. But without success. I tried to design a test case by benchmarking how we did it with the DVL sensor. I've made a urdf xacro file to describe the multibeam sonar sensor and attached it in the robot xacro file before default_joint_stat_publisher is called. But I get errors on Rviz when visualizing the point cloud data. I could really use some advice or help on how to tackle this problem. @bsb808 @dtdavi1 @crvogt Quickstart
roslaunch uuv_dave uuv_dave_multibeam_sonar.launch |
@woensug-choi Investigating... |
@woensug-choi FYI I'm just making a standalone MBS example and from there figure out what's going on when added to the robot model :) |
@woensug-choi I noticed that by using the MBS plugin alone the point cloud in RVIZ reflects what @lauralindzey mentioned. I added a |
@crvogt What do we need to do to close this issue? Adding |
@woensug-choi Sorry, just saw this. I'll make a PR for this. |
bump =) We have another project that we're ramping up with the simulator, and this issue just came up again. Ideally, the fix would be for the sensor used to generate the image_raw/image_depth data to be properly aligned, but I think a hacky solution could be to just switch the order in which beams are added to the message. (ln 843 of gazebo_ros_multibeam_sonar.cpp) I'd be happy to hop on a Zoom if that would be useful for debugging. |
@lauralindzey @woensug-choi Thanks, bump noted. Have you tried the |
@lauralindzey I am not much familiar with the subject here. Could you try the solution @crvogt suggests? If it did not work, let's arrange a meeting :) Example : |
I'm seeing two different issues with the frames in the Gazebo plugin:
If I attach the sensor to a link where the axes are [forward, right, down] (as specified in the definition of acoustic_msgs/SonarImage), the sensor_msgs/Image that is published will have the correct field of view but the image will be flipped left-right when viewed in image_view.
If we try to visualize the published pointcloud, it has the correct geometry for the sensor's field of view, but has been pitched down by 90 and then rolled by 90.
The text was updated successfully, but these errors were encountered: