slam_toolbox and LIDAR: error discarding message because the queue is full

[sebastianfer]

Hi, I am trying to generate a map using LIDAR, and I am following the instructions listed here: https://github.com/iRobotEducation/create3_examples/tree/e0dcd78f9b00c6fa1533d1d4cb37542c29095edc/create3_lidar

My Raspberry Pi 4, robot, and virtual machine are all in the same network, and I am getting full topic list. I opened three terminals and logged into the Pi for the setup. The command that starts the sensors seems to be working fine (ros2 launch create3_lidar sensors_launch.py), but when I run slam toolbox in the second terminal (ros2 launch create3_lidar slam_toolbox_launch.py), I get the following message:[async_slam_toolbox_node-1] [INFO] [1658262909.942375981] [slam_toolbox]: Message Filter dropping message: frame 'laser_frame' at time 1658262909.862 for reason 'discarding message because the queue is full'

Any ideas how I can solve this?

[justinIRBT]

Is your Create3 on the network? This error could be because your robot's system time doesn't match that of the computer your lidar is attached to. If your robot is on the network, it will use NTP to sync to global time. Alternatively, if you don't want to / can't get your Create3 on the network, you can sync the time between your SBC and Create3 following the instructions here:
https://iroboteducation.github.io/create3_docs/setup/compute-ntp/

[justinIRBT]

Does the tf tree show up properly on rviz2? Is you drive the robot and select the odom frame in rviz, is the laser frame moving with the robot if you display the tf tree?

[justinIRBT]

https://answers.ros.org/question/389383/slam_toolbox-message-filter-dropping-message-for-reason-discarding-message-because-the-queue-is-full/

[sebastianfer]

I did not move the robot yet since I thought the map would not be generated if that problem was not solved. Could you tell me where I can find the tf tree in the workspace and how can I change the parameters?

[justinIRBT]

The tf tree publishes on /tf. In rviz2 you can add a viewer for TF that will show you the tree. In the example, this line publishes the link between the robot tree and the lidar
https://github.com/iRobotEducation/create3_examples/blob/galactic/create3_lidar/launch/sensors_launch.py#L12

[sebastianfer]

So, I ran the command again today. I couldn't display the tf tree in rviz, but I did echo of the topic on Terminal. This is the last thing it returned:

transforms:

• header:
  stamp:
  sec: 1658412365
  nanosec: 257626853
  frame_id: odom
  child_frame_id: base_link
  transform:
  translation:
  x: -2.061793088912964
  y: 3.644369125366211
  z: 0.08317188918590546
  rotation:
  x: -0.007724330760538578
  y: -0.000545671209692955
  z: 0.81966233253479
  w: 0.5727946758270264
• header:
  stamp:
  sec: 1658412365
  nanosec: 257626853
  frame_id: odom
  child_frame_id: base_footprint
  transform:
  translation:
  x: -2.0623789359229407
  y: 3.643689893423769
  z: 0.0
  rotation:
  x: 0.0
  y: 0.0
  z: 0.8196700811386108
  w: 0.5728358626365662

---

Does this mean it is doing the transformations properly?

[sebastianfer]

The solution to this problem can be found in #118. After fixing that, my Lidar was detected and registered.