R2300 can't visualize point cloud in rviz about pf_lidar_ros_driver

[wsaihopfsg]

Both /cloud & /cloud_1 are not publishing. In fact all topics are not publishing except for /tf, /r2300_node/parameter_descriptions & /r2300_node/parameter_updates

The issue is quite confirmed to be with WSL2, since in the initial request_handle_udp and subsequent start_scanoutput are TCP-based and initiated by the driver within WSL2. (WSL2 -> R2300)

When the UDP response comes in the other direction (R2300 -> WSL2), it doesn't work. The official work-around involves proxyport forwarding with netsh, but it doesn't work for UDP.

Following the steps here I have managed to get the rviz visualization. Nothing is wrong with the driver, just need to massage some WSL2 settings.

[john-sokko]

Here is a screenshot.

[hsd-dev]

Is this on Ubuntu 20.04 and ROS Noetic? I am having the same problem on Noetic.

[john-sokko]

Yes.

[hsd-dev]

looks like it is a known error: https://answers.ros.org/question/355595/rviz-tf-error-failed-to-transform-from-frame-map-to-frame-map/

[wsaihopfsg]

I can successfully invoke roslaunch pf_driver r2300.launch, but cannot see anything in rviz.

... logging to /home/pepperlfuchsctss/.ros/log/c02be5fa-3a02-11eb-9f3e-00155d97d3b7/roslaunch-1806PTN523-2997.log
Checking log directory for disk usage. This may take a while.
Press Ctrl-C to interrupt
Done checking log file disk usage. Usage is <1GB.

started roslaunch server http://1806PTN523:37105/

SUMMARY
========

PARAMETERS
 * /rosdistro: melodic
 * /rosversion: 1.14.10

NODES
  /
    r2300_node (pf_driver/ros_main)
    ring1 (tf/static_transform_publisher)
    ring2 (tf/static_transform_publisher)
    ring3 (tf/static_transform_publisher)
    ring4 (tf/static_transform_publisher)
    rviz (rviz/rviz)

ROS_MASTER_URI=http://localhost:11311

process[r2300_node-1]: started with pid [3012]
process[ring1-2]: started with pid [3013]
process[ring2-3]: started with pid [3014]
process[ring3-4]: started with pid [3015]
process[ring4-5]: started with pid [3025]
[ INFO] [1607515719.099464700]: Device found: OMD10M-R2300-B23-V1V1D-4S
process[rviz-6]: started with pid [3037]
[ INFO] [1607515719.173835000]: Device state changed to Initialized
QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-pepperlfuchsctss'
[ INFO] [1607515719.208733100]: Device state changed to Running

I checked all the cloud related topics /cloud_1 /cloud_2 /cloud_3 /cloud_4, all are not publishing.

The only topic that seems to be actively publishing is /tf.

The unit is not defective since I can visualize using another P+F software.

I am running WSL with Ubuntu 18.04 and ROS melodic.

Thank you.

[hsd-dev]

@wsaihopfsg check the link posted in this thread. It is apparently a known issue in Noetic.

[wsaihopfsg]

@ipa-hsd Sorry I made a typo, I am using melodic

[hsd-dev]

It would be difficult to replicate this error since there has not been an issue with Melodic before and there seems to be no error output in your case. Can you check the output of the pointcloud and laserscan on terminal with

rostopic echo /cloud
rostopic echo /cloud_1