Deformable Bimanual Manipulation Benchmark
Collection
Bimanual cloth and bag manipulation in simulation: LeRobot v3.0 datasets and the policies trained on them. • 7 items • Updated
video video 249 1.41k |
|---|
A dual-arm Piper flattens a crumpled t-shirt on a table. Success combines a coverage signal (catches 'still crumpled') with a shape-distance signal against a canonical flat pose (catches 'spread out but still folded').
Part of a bimanual deformable-object (cloth / bag) manipulation benchmark. Physics is a GPU cloth solver co-simulated with the robot in a single model; observations are rendered with a photorealistic renderer.
| Format | LeRobot v3.0 (parquet + mp4, compacted) |
| Robot | sim_dual_piper_deformable |
| Episodes | 200 |
| Frames | 41,464 |
| Tasks | 1 |
| Control rate | 25 fps |
| Cameras | static_cam, right_hand_cam, left_hand_cam |
| Video | 1280×720, H264 |
| State / action | 14-D each (left 6 joints + gripper, right 6 joints + gripper) |
from lerobot.datasets.lerobot_dataset import LeRobotDataset
ds = LeRobotDataset("hwk0809/flatten-tshirt")
If your torchcodec build cannot decode these videos, pass video_backend="pyav".
Apache-2.0.