Build a Robot Pick-and-Place Simulation with SimReady Assets
Reported by Hugging Face and ecosistemastartup2 sources ↓
Hugging Face published a walkthrough using a UR17G arm, Robotiq 2F-140 gripper and beverage can in Isaac Sim, with an AI agent helping assemble the scene and generate controls for a pick-and-place task.
Key points
- The task has the robot grasp, lift, move and release a beverage can
- The arm and gripper validation reports pass PhysX driven-joint checks
- The example uses GPT-6 Sol; other agent and model combinations may work
- The walkthrough calls for checking joints, movement and contact before the task
Sources
Details
The walkthrough uses SimReady assets, which include simulation properties such as collisions, mass, joints and joint drives. Hugging Face says the assets’ validation reports provide evidence that they meet specified profiles; the arm uses Robot-Body 4.0.0 and the gripper uses Robot-Gripper 3.0.0.
Users download the three asset packages, assemble the arm and gripper in Isaac Sim, add a work surface and starting positions, then run controls for the arm and gripper. The example uses GPT-6 Sol to help generate and run task code, but Hugging Face cautions that generated code and results can vary and should be inspected.
The walkthrough recommends checking joint settings, movement and contact behavior before attempting the pick-and-place run. The press article reports that the example does not include vision or replanning and does not establish performance with uncertain object positions or on physical hardware.
Written from coverage as of
Images
- Hugging Face
- ecosistemastartup
- Hugging Face