Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Question] How to grasp and lift an object with a Franka robot in a Direct RL approach? #1270

Open
pmfaustino opened this issue Oct 18, 2024 · 2 comments

Comments

@pmfaustino
Copy link

I have seen and tried out the Isaac-Lift-Cube-Franka-v0 Environment, where the Franka robot is able to pick up and lift a cube following a Manager Based RL approach and I was trying to obtain similar results following a Direct RL approach instead. Using the Isaac-Franka-Cabinet-Direct-v0 Environment as a basis, a few changes in the code have been made to be able to reach the cube and lift it. The issue is that the robot gets close to the cube, but doesn't grasp it or lift it and I can't understand what I'm missing/doing wrong. You may find the code attached below. Any help would be appreciated.

franka_cuboid_env.txt

@mpgussert
Copy link
Collaborator

Hello @pmfaustino!

Unfortunately, there's not a lot we can do to help without more information. Can you provide a code snippet or example?

Also, How are you controlling the robot in both cases? Are you using the "cube grabbing policy" from the first environment, in the second?

thanks!
Gus

@pmfaustino
Copy link
Author

Hello @pmfaustino!

Unfortunately, there's not a lot we can do to help without more information. Can you provide a code snippet or example?

Also, How are you controlling the robot in both cases? Are you using the "cube grabbing policy" from the first environment, in the second?

thanks! Gus

I apologize for the misunderstanding. The code that I have been running is the one from the attached file. It is a python file, but I was unable to attach it with the proper extension due to GitHub's policy for file attachments.

But that's basically the gist of it, I was trying to achieve similar results to the first environment (Manager Based RL approach) by altering the code from the second environment (Direct RL approach). Have not been able to successfully do so though.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants