record-0528-2acts-50eps
收藏Hugging Face2026-05-28 更新2026-05-28 收录
下载链接:
https://huggingface.co/datasets/Panda512/record-0528-2acts-50eps
下载链接
链接失效反馈官方服务:
资源简介:
该数据集由LeRobot创建,专注于机器人技术领域,包含51个episodes,总计80475帧数据,存储为parquet格式,并附带mp4格式的视频文件。它记录了so_follower类型机器人的动作和观察数据:动作数据包括肩部平移、肩部升降、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置等6个浮点数值;观察数据包括相同的6维机器人状态,以及来自上方和前方两个视角的图像数据,图像分辨率为480x640,3通道,帧率为30fps。数据集用于训练和评估机器人控制模型,支持从多模态输入中学习策略。
This dataset was developed by LeRobot and focuses on the field of robotics. It contains 51 episodes with a total of 80,475 frames, stored in Parquet format, and is accompanied by MP4-format video files. It records the action and observation data of the so_follower-type robot: the action data includes 6 floating-point values, namely shoulder translation, shoulder lift, elbow flexion, wrist flexion, wrist rotation, and gripper position; the observation data includes the same 6-dimensional robot state, as well as image data from two perspectives: overhead and front-facing views. The images have a resolution of 480 × 640, 3 color channels, and a frame rate of 30 fps. This dataset is utilized for training and evaluating robot control models, supporting policy learning from multimodal inputs.
提供机构:
Panda512创建时间:
2026-05-28
原始信息汇总
数据集概述
该数据集名为 record-0528-2acts-50eps,由 LeRobot 创建,采用 Apache-2.0 许可证。
- 机器人类型:so_follower
- 总片段数:51
- 总帧数:80475
- 总任务数:1
- 数据文件大小:100 MB
- 视频文件大小:200 MB
- 帧率:30 FPS
数据集结构
- 分割方式:所有 51 个片段均作为训练集(
train: "0:51")。 - 数据格式:
- 数据文件路径:
data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet - 视频文件路径:
videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4
- 数据文件路径:
特征说明
数据集包含以下特征:
| 特征名 | 数据类型 | 形状 | 字段说明 |
|---|---|---|---|
action |
float32 | (6,) | 动作指令,包含 6 个关节位置(shoulder_pan, shoulder_lift, elbow_flex, wrist_flex, wrist_roll, gripper) |
observation.state |
float32 | (6,) | 观测状态,同样为 6 个关节位置 |
observation.images.up |
video | (480, 640, 3) | 上方摄像头图像,尺寸 480x640,3 通道,AV1 编码,30 FPS |
observation.images.front |
video | (480, 640, 3) | 前方摄像头图像,尺寸 480x640,3 通道,AV1 编码,30 FPS |
timestamp |
float32 | (1,) | 时间戳 |
frame_index |
int64 | (1,) | 帧索引 |
episode_index |
int64 | (1,) | 片段索引 |
index |
int64 | (1,) | 全局索引 |
task_index |
int64 | (1,) | 任务索引 |



