1
$\begingroup$

My robot consists of an IMU, a LiDAR and a TOF sensor(in form of depth camera), which are all connected to the base_link. I want to visualize all the sensor data and the end goal is to achieve sensor fusion.

When I added the topics in RVIZ2, it visualizes only the point cloud from the LiDAR(red cylinder) while dropping the messages from other sensors. I can not see any data from the depth camera(blue square)

[rviz2-10] [INFO] [1712592704.444317657] [rviz]: Message Filter dropping message: frame 'robot/base_link/depth_camera' at time 410.900 for reason 'discarding message because the queue is full'

[rviz2-10] [INFO] [1712592704.506784400] [rviz]: Message Filter dropping message: frame 'robot/base_link/imu_sensor' at time 402.000 for reason 'discarding message because the queue is full'

I have to change the fixed frame to robot/base_link/depth_camera to visualize the depth camera message(which is connected to the TOF sensor link) but then there will be no data from the lidar, and the robot model becomes unavailable. So my problem is, how to visualize the point cloud from both the lidar and depth camera sensors with the fixed frame set to "world" so that I can see the data from the depth camera along with my robot model?

The TF2 frames looks like this RQT_graph which all seem to be correct.

Edit: the TF tree in rviz

Source Code(I have tried changing frame_name to frameName and ignition_frame_id but none works):

$\endgroup$

1 Answer 1

0
$\begingroup$

I presume that the sensors are simulated with gazebo. I think there is something wrong with the tf of the robot. In the gazebo wiki for the depthcam you can change the frame name. I think if you change the frame name to tof_sensor_link_1 it should work.

$\endgroup$
1
  • $\begingroup$ Hi, thanks for your answer, I have added the 'ignition_frame_id' to the sensor, and the tf seems correct, however it still gives the same results. I will attach the source code for clarification $\endgroup$
    – Bozzer
    Commented Apr 11 at 18:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.