0
$\begingroup$

Good morning, I open this question as I have not found a way to get the model_entity that collides with the gpu_lidar sensor.

I would like to be able to attach a gripper with such entity, but I can't know its name and location a priori. The only info I have is the relative distance from the lidar sensor msg.

Is there any way to get the first mesh to collide with the ray of the gpu_lidar and subsequently trace back to the link to create a detachable joint?

Improve this question
$\endgroup$
1

1 Answer 1

Reset to default
1
$\begingroup$

From your question at Gazebo Answers I understand that you want to move the gripper to the detected lidar position, and then grasp the object by making a detachable joint. But for this you need the entity ID of the detected object. Correct?

If so, you can probably use a PerformerDetector system with the PerformerDetector's region attached to your gripper. It publishes the name and the entity ID.

See this documentation and the PerformerDetector example world.

Improve this answer
$\endgroup$
3
  • 1
    $\begingroup$ One caveat is that the PerformerDetector uses axis aligned boxes, so it may not work well in some cases. Another option to look at is logical cameras $\endgroup$
    – azeey
    Commented Aug 9, 2023 at 20:00
  • $\begingroup$ Thanks a lot for the reference! The difficulty really lies in making attachments with an entity whose location and ID I cannot know a priori, I would like to emulate the most realistic behavior possible and looking on the net I found only about vacuum gripper (not working) and I was looking for something that would narrow down the check on entities in a particular area of the gripper, and then make attachments with the closest one based on the lidar msg. $\endgroup$
    – Gabriele
    Commented Aug 10, 2023 at 12:58
  • $\begingroup$ There are other possible solutions, e.g. transform the laser scanner object points into world coordinates, and compare these to object coordinates from a pose publisher to discriminate which objects corresponds to the laser scanner detection. But as @Tully already wrote in another comment: you will need to give more detailed info on what specific you try to achieve (preferably with a sketch or image), how exactly you are currently trying, and what specific problems you run into that make it not work. $\endgroup$
    – JRTG
    Commented Aug 10, 2023 at 13:12

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.