ros2
Datasets
All datasets matching “ros2”eval_ROS2SmolVLAThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "ros2",
"total_episodes": 244,
"total_frames": 333872,
"total_tasks": 5,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:244"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/una-auxme/eval_ROS2SmolVLA.ROS2SmolVLA_ur10e_no_joints_crop_pick_placeThis is the training dataset for our ROS2SmolVLA project
It was recorded by teleoperation of our UR10e lightweight industrial robot through our ROS2SmolVLA setup utilizing LeRobot.
The action space is:
"linear_x.vel",
"linear_y.vel",
"linear_z.vel",
"angular_x.vel",
"angular_y.vel",
"angular_z.vel",
"gripper.pos"
The observation space is:
"pose.x",
"pose.y",
"pose.z",
"pose.quat_x",
"pose.quat_y",
"pose.quat_z",
"pose.quat_w"
one 720x720 and two 1280x720 camera streams.… See the full description on the dataset page: https://huggingface.co/datasets/una-auxme/ROS2SmolVLA_ur10e_no_joints_crop_pick_place.ros2-benchmark-configstest_ros2_cameraThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "so100",
"total_episodes": 1,
"total_frames": 298,
"total_tasks": 1,
"total_videos": 2,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:1"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/yadunund/test_ros2_camera.ros2_tms_for_constructionpiper_dual_stack_cups_ros2_shifted