ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question
0

MoveIT! autonomous control UR5

asked 2019-01-31 06:20:05 -0500

sisko gravatar image

updated 2019-01-31 06:41:46 -0500

Hi all.

I have a task to use a UR5 robot to track QR codes on moving turtlebot3 robots.

My research led me to the MoveIT! package but everything I've seen goes through the moveit setup process and stops at the planning and execution interface on RVIZ.

If MoveIT! can allow a UR5 to function autonomously, I need direction on how to use MoveIT! on a real robot. My task requires my real UR5 robot to act autonomously and identify a specific turtlebot by it's QR code using a camera end effector.

I would appreciate any help.

Update:

By "function autonomously" I mean how can I utilise MoveIT! so that any time my UR5 "sees" a target QR code, MoveIT! moves the end effector towards the target.

As I mentioned previously, the tutorials and book I have been referring to stop at manually using Rviz to plan and execute movements. I would like the robot to plan and execute movements autonomously based on a topic I will subscribe to which will be responsible for identifying the QR code.

edit retag flag offensive close merge delete

Comments

What do you mean by 'function autonomously'? MoveIt can produce motion plans and execute them, but it doesn't decide what needs to be done in the first place. If you can explain the type of behaviour you want we may be able to help you.

PeteBlackerThe3rd gravatar image PeteBlackerThe3rd  ( 2019-01-31 06:27:18 -0500 )edit

I added an update. I hope it clarifies things.

sisko gravatar image sisko  ( 2019-01-31 06:42:36 -0500 )edit

1 Answer

Sort by ยป oldest newest most voted
0

answered 2019-01-31 07:02:27 -0500

What you want to do is outside the scope of MoveIt, although a solution may well make use of MoveIt along with other packages to achieve the final goal.

If you can calibrate the end effector camera, setup a node to detect the poses of QR codes in camera images, then you have all the required information to send an EE pose goal to moveit, and in a rather jerky way achieve what you want.

But you need to define a bit more clearly how you want the robot to achieve the required behaviour and very importantly how it can do it safely without colliding with any static or mobile objects in its work-cell.

edit flag offensive delete link more

Question Tools

2 followers

Stats

Asked: 2019-01-31 06:20:05 -0500

Seen: 319 times

Last updated: Jan 31 '19