[sam.choy]

Hello roboDK team,

my request is about the implementation of a driver and a post processor to control the robot Dobot MG400. In the roboDK library already exists the robot, which allows to perform simulations.

It would be interesting to connect the real robot to roboDK.

这是各种Dobot我的链接ndustrial robot:
https://github.com/Dobot-Arm/TCP-IP-Prot...ADME-EN.md
The SDK is suitable for the secondary development of Dobot collaborative robot CR/MG400/M1Pro series. Based on TCP / IP communication protocol, and provides the support of Java / C++ / C# / Python / ROS / MATLAB / LabVIEW / Android Demo.

In our lab we already purchase 10 sets of DoBot MG400 and 31 sets of RoboDK licenses.
we would be interested in testing the driver and helping with the development.

Best regards,
Sam Choy

[Albert]

We currently have a driver for Dobot Magician robot. Do you know if it works on the Dobot MG400 as well?

If not, we would be happy to work with you to improve our Dobot driver to support the MG400 robot. It is better if you contact us by email at info@www.sinclairbody.com so we can start working on this.

[sam.choy]

We currently have a driver for Dobot Magician robot. Do you know if it works on the Dobot MG400 as well?

If not, we would be happy to work with you to improve our Dobot driver to support the MG400 robot. It is better if you contact us by email at info@www.sinclairbody.com so we can start working on this.

Dobot Magician post-processor cannot apply on Dobot MG400, since it uses Lua script.
I made the first version post-processor and it seems not to be working very well currently.

[Albert]

Great, thank you for sharing.

What issues did you have with this post processor? If you can share the error messages or error codes you get we can help you better.

[Albert]

I noticed there was an issue when generating joint moves. You should provide joint values, not the pose. I attached the post processor with some minor edits.

[sam.choy]

How ever I found the joint move should provide cartessian point instead of joint value.
I think the joint value should apply in .

[Albert]

With the information you provided it may be more robust to use an absolute joint movement for joint moves (the MoveJ function in RoboDK post processor)

Also, it may be better to not specify the arm orientation for linear movements.

I attached the updated post processor.