Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Camera robot detection possibility?
#1
Lightbulb 
Hi,
I'm trying to code a program for my robotic system, that detects holes in a piece, and have the robot move towards the hole and have the gripper( with camera mounted on it) to be exactly above the hole , so an image to be taken? Any directions or tips?

Thanks!
#2
I attached an image of the set up.
The intent is to to have the robot move , until the center of one of the outer circles shown is perfectly on the principle axis of the camera?
image link : https://ibb.co/3Y2w33N 


I'm using the blob detection example script for detection : https://robodk.com/doc/en/PythonAPI/v0/e...-detection
#3
Hello Nour,

Is this for part inspection or part alignment?
Are the holes randomly placed?

If this is for inspection and the hole locations are known, it is better to import your 3D model in RoboDK and generate a path moving over them.
If you really need to be perfectly aligned, you can then adjust the XY position of the TCP based on the offset observed with the camera.

For instance:
  • Move the robot to the first hole, as defined in your model
  • Take a capture, find the blobs
  • Find the centre of the blob closest to the centre of the image, and it's offset from the centre of the camera
  • Move the tool by half that offset and take another capture
  • Do this until you are within tolerance
Take a look at our sample scripts in our Python documentation:
Find useful information about RoboDK and its features by visiting our Online Documentation and by watching tutorials on our YouTube Channel.
#4
Hi, I was actually able to do that thanks to you!, my current consideration is that I was trying to change the detection method using hough circle transform, any guidance regarding that?
Thank you
Nour
#5
Hi Nour,

OpenCV has a great tutorial on this method: https://docs.opencv.org/3.4/d4/d70/tutor...ircle.html
Find useful information about RoboDK and its features by visiting our Online Documentation and by watching tutorials on our YouTube Channel.
  




Users browsing this thread:
1 Guest(s)