Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5

Robot Holds Object

#1
Hello Everybody,

I'm pretty new to RDK, but I've followed the professional training.

I want to follow the blue path below in the picture:

   

In my program, the robot does something weird in the simulation after frame 280.

Also, I'm running into singularities (Noticeable in the first and last couple of frames). Any tips for a better path?

The station is in the attachments.

Thanks a lot!


Attached Files
.rdk   Help Project.rdk (Size: 1.15 MB / Downloads: 36)
#2
You can reduce the pitch between curve points and try to find the tool's rotation around the Z axis (and fix it if possible) when the robot arm is far away from the singularities:

   

I attached a modified station


Attached Files
.rdk   Help Project_md.rdk (Size: 1.2 MB / Downloads: 32)
#3
(03-06-2024, 09:09 AM)Sergei Wrote: You can reduce the pitch between curve points and try to find the tool's rotation around the Z axis (and fix it if possible) when the robot arm is far away from the singularities:



I attached a modified station

Hello Sergei,

Thank you so much for this station; it works like a charm!

But I would still like to know how to get these settings if I had to do it myself.

For my understandings:

1) Reducing the Pitch: This adjusts the length between two points. The closer the points are, the better. Or is this just for accuracy?

2) Finding the right tool orientation: I can see the adjustments in the preferred tool path. How can I determine the values you put in? They seem quite specific.

Thanks Again!
#4
1. Due to the specifics of the 'robot holds object' algorithm, this allows you to explicitly set the nodes for the trajectory and increase the accuracy of following it.

2. The method may be different in each case.
For wrist singularities, j5 = 0 is typical.
I chose one of the approach points where j5 ~ 90 degrees and used the teach button to record the values, then I reduced the allowable tool Z rotation as much as possible and set a low value for pitch change.
This allowed us to avoid crossing the j5 zero mark while moving along the trajectory.
#5
Wow, Yes this totaly works great for me.

So let's say I have a different singularity that I cross. If I teach it so its far away from the singularity, the robots path should adjust its won't cross it?

Thank you so much for this already!
#6
Quote:So let's say I have a different singularity that I cross. If I teach it so its far away from the singularity, the robots path should adjust its won't cross it?

I'd say it's not a solution for a general case, but it can be successful for examples like you shared.
#7
Allright, thanks for sharing the information!

Helped me a lot with my project, making good steps.

Thanks again!
  




Users browsing this thread:
1 Guest(s)