Hello,
I’m new to this forum. I have recently been working with the universal robot and it has been a great experience so far. I now received the intel realsense depth camera D435i. My goal at my university is to integrate this camera on the ur3e and try to program a vision system that can make the robot do pick and place operations on predefined objects.
I’m doing some research on the topic but there are so many programs and ways to do it, so I actually don’t know where to start. I hope this question is not to wide to answer but I thought maybe someone knows which programs are compatible to combine this camera with the UR3e. Or just a source where I can read more about it would also be great.
Thank you in advance!
Cas
Hi,
I actually had a similar project for my dissertation, with an UR5e and the camera you mentioned. I used Python and socket TCP/IP communication to send the information to the robot.
I hope this helped you.
Greetings
Pedro Monforte
Hello Mr. Pedro!
I’m from Brazil and i’m developing the same project that you mentioned here, for my “Final Paper” of Bacharelor in Electrical Engineering (UDESC: Universidade do Estado de Santa Catarina).
However, I’m in the first steps of this project, so if you have some tips, can you please tell me?
Can we trade our emails to talk with each other/share the goods and bads about this developing?
For now my “chosen path” was create a socket connection by IP (TCP-IP) with the robot, cuz i searched a lot about another ways and the most include:
(1) A Intel Camera D435i (cuz this camera has 3D compatibility) Problem: I don’t have funds available for now in my University for this project - I changed this for a Common Logitech Webcam.
My goal is do first-things-first, so for this Final Paper I want to just send the coordinates to the robot and let the UR built-in Kinematics choose the path.
(2) using the tools of Extended Robotics (that isn’t free). Problem: My university requires that our internal projects use Open-Source tools.
So the path using TCP-IP was no-money driving and for me is the best way.
Thanks a lot!
Hi, I am trying to do a similar task with my UR5e and the D435i, can you give me some information as to how you’re going about doing this task? I have gone down the socket TCP/IP method to communicate with the robot, but I am running into multiple issues (I can’t send more than 1 move command to the robot with the moveL/moveJ commands). My end goal is to be able to detect predefined objects with the D435i and then pass that pose information to the python program.
My initial step is to get the UR5e to move to 3-4 target poses and then do something with my Hand E Gripper, if anyone has some sample code that I can work off of, or some advice/resources where I can get some help with this, it’d be extremely helpful!
Thanks, and I’d love to have discussions about this further.