USC Cloth Sim is a dataset developed by USC's RESL Lab for deformable object manipulation research. It contains 800 episodes of a Franka Emika Panda robot interacting with cloth materials, including RGB images, depth data, and robot joint states. The dataset supports research in cloth folding, smoothing, and dynamic manipulation, with a focus on learning from expert demonstrations. It is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) license, allowing free use and modification for academic and commercial purposes. USC Cloth Sim is accompanied by evaluation metrics and simulation environments, making it suitable for studying deformable object manipulation in robotics.
The robot manipulates a deformable object (cloth on a tabletop) along a diagonal.
| Field | Value |
|---|---|
| Action Space | EEF Position |
| Control Frequency | 10 |
| Depth Cams | 0 |
| Episodes | 170 |
| File Size | 0.08 GB |
| Gripper | null: Robot agnostic sim data |
| Has Camera Calibration | False |
| Has Proprioception | False |
| Has Suboptimal | False |
| Rgb Cams | 1 |
| Robot | Sawyer |
| Robot Morphology | Single Arm |
| Scene Type | Table Top, Kitchen (also toy kitchen) |
| Wrist Cams | 0 |