Hi,
I’m trying to use RL to train pick and place robotic task. For this, I have a robot arm and a cube. At reset, the cube should start in the same position, I’m trying to train a single agent with multiple environments.
I’m using a GeometryPrimView.set_world_poses to move the cubes to the default position.
Am I duoing this right or should use other method?
Hi there, the GeometryPrimView class can be used for this use case. To speed up performance, we also provide vectorized APIs in RigidPrimView.set_world_poses, which can set the transforms of multiple rigid bodies at the same time.
Thanks, that works.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.