0
$\begingroup$

I have a UR5e installed with a simple camera on its end effector. The robot is at a known position (Cartesian pose/angles and joint angles)

Using the camera I took a picture of a QR code and got the object's real life coordinates to the camera using OpenCV's SolvePNP. This gives me a 3x1 translation vector and a 3x1 rotation vector.

I'd like to tell the robot to move to that position in translation and angles.

I guess first step would be to inverse that result so I have the QR code's position from the camera's axes point of view.

How do I "add" this movement to the robot's arm?

(Using Python and ut_rtde)

Thank you

$\endgroup$

2 Answers 2

0
$\begingroup$

Just run some inverse kinematics and get the 6 joint angles that move the camera to the wanted position.

And find the example Python code that controls each motor angle of the robot and give your 6 values above.

$\endgroup$
0
$\begingroup$

If extrinsic parameters of camera (Transformation between end-effector and camera) are known, you know the transformation matrix between camera and robot base. So, definitely you know the T matrix between base and QR code in cartesian space, so you coulduse a function to move robot respect base using cartesian movement or by joints (calculating IK knowing T matrix between base and QR)

$\endgroup$

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.