Mitsubishi RV-6SDL Robot

I've also had a look at the trial version of robodk:

robodk.com/simulation

It also has a plugin for solidworks to extract contours directly from the CAD. However this obviously isn't exactly cheap and it would still require a post processor for LCNC gcode but for a commercial application it might still be worth looking at.

[edit]
Just realized you guys already had this conversation. forum.linuxcnc.org/30-cnc-machines/37627...otos?start=20#148430

Could this be done with Linuxcnc :

tommylight wrote:

And they sold B.D. !!!!!!!
For a measly 1B !
How the hell is Apple worth nearly 2T while BD only 1B???

The robots are slowly rising to power to avoid suspicion? With the rise of AI these robots will be making us their slaves.

Back on topic, do you guys see any reason to use ROS? I have been trying to get my head around what it does for a while but I cannot see what it achieves that LinuxCNC cannot.

They talk about obstacal avoidance and such. I guess thats useful if your not using a CAM program to program the robot. I also have been looking at the Scan-N-Plan feature. It would be great if the robot could scan the work piece and come up with its own movement paths. Dont know if that is possible yet or not or how well it works outside of a lab environment.

do you guys see any reason to use ROS?

That is a good question. I don't think I know enough about ROS to answer that. In my opinion for parts handling and machine tending you don't need ROS and for welding you need path blending through a trajectory planner.
If you are a good programmer I'm sure a framework like ROS could bring a lot of new possibilities.
Personally I don't think automatic feature recognition with path planning in 3D for a 6 DOF robot is within reach (2D might be a different story of course). Extracting a 3D contour from a CAD model into a CAM and then using a postprocessor to create a program is the standard and I would focus on that. In the end it will be a program with a large number of G01 moves that need to be blended to a constant motion. My Mitsubishi controllers made in 2013 can't do that but more modern ones can. What you need for that is a trajectory planner I don't see how ROS would be of any help there.
What ROS might be able to to is help with the tool orientation say to keep it at a certain angle to the path but I think it will be more economic to invest in something like RoboDK that is already made up to do that instead of trying to program that all again in ROS.

[edit]
If I was trying to build a welding robot I would probably focus on RoboDK and see if it can be made to spit out Gcode that can be used with linuxcnc.

Yeah I think for welding items like columns with cleats and such RoboDK will be the main tool I would need to generate my programs. But maybe ROS might be good for bin picking jobs where there is no set item to be picked.

Hopefully I have most of my projects finished by the middle of next year so I can start getting into the software side of robotics. I brought a copy of RoboDK a few months ago and Albert has created 3d models of all of my robots for me. Its going to be fun.

I brought a copy of RoboDK a few months ago and Albert has created 3d models of all of my robots for me.

Interesting. Has he created a postprocessor that spits out gcode?

[edit]

And yes for bin picking I can see ROS being potentially very useful but there have been some who have successfully incorporated OpenCV directly into LCNC.