my_openarm_dataset_m8
收藏资源简介:
该数据集是一个基于LeRobot工具创建的机器人控制数据集,专为机器人学习和模仿任务设计。数据集包含80个episodes,总计41280帧,数据文件大小为100MB,视频文件大小为200MB。数据记录了openarm_follower机器人的动作和观测信息:动作包括8个关节位置和夹爪位置;观测状态与动作相同;视频观测包括胸部摄像头(分辨率848x480,30fps)和右腕摄像头(分辨率640x480,30fps)的RGB视频流。此外,还包含时间戳、帧索引、episode索引、任务索引等元数据。数据以Parquet格式存储,视频以MP4格式存储,适用于机器人控制算法的训练和评估。
This dataset is a robot control dataset created using the LeRobot tool, specifically designed for robot learning and imitation tasks. It contains 80 episodes, totaling 41,280 frames, with a data file size of 100MB and a video file size of 200MB. The dataset records actions and observations of an openarm_follower robot: actions include 8 joint positions and gripper position; observation states are the same as actions; video observations include RGB video streams from a chest camera (resolution 848x480, 30fps) and a right wrist camera (resolution 640x480, 30fps). Additionally, it includes metadata such as timestamps, frame indices, episode indices, and task indices. Data is stored in Parquet format, and videos are stored in MP4 format, suitable for training and evaluating robot control algorithms.
数据集概述:Nick3421/my_openarm_dataset_m8
- 许可证: Apache-2.0
- 任务类别: 机器人学
- 创建工具: LeRobot
数据集结构
- 数据集版本: v3.0
- 帧率: 30 FPS
- 总片段数: 80
- 总帧数: 41,280
- 总任务数: 1
- 机器人类型: openarm_follower
- 数据分割: 训练集(80个片段,对应索引 0:80)
数据特征
| 特征名称 | 数据类型 | 形状 | 说明 |
|---|---|---|---|
action |
float32 | [8] | 包含7个关节位置和1个夹爪位置 |
observation.state |
float32 | [8] | 与action相同,表示7个关节位置和1个夹爪位置 |
observation.images.chest |
video | [480, 848, 3] | 胸部摄像头图像,分辨率 480×848,AV1编码,30 FPS |
observation.images.right_wrist |
video | [480, 640, 3] | 右手腕摄像头图像,分辨率 480×640,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] | 索引 |
task_index |
int64 | [1] | 任务索引 |
数据文件
- 数据存储路径:
data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet - 视频存储路径:
videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4 - 数据文件大小: 约 100 MB
- 视频文件大小: 约 200 MB
- 分块大小: 1000帧/块
可视化
该数据集支持通过 LeRobot 可视化工具 进行在线浏览。



