rollout_pick_n_place_dagger_r7
收藏Hugging Face2026-06-06 更新2026-06-06 收录
下载链接:
https://huggingface.co/datasets/anikitakis/rollout_pick_n_place_dagger_r7
下载链接
链接失效反馈官方服务:
资源简介:
该数据集使用LeRobot创建,是一个专注于拾取和放置任务的机器人学数据集。它包含动作和观察数据:动作数据由6个关节位置组成,包括肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置;观察数据包括相同的关节位置状态,以及前视图像和注释图像(以视频格式存储,分辨率为480x640,3通道,帧率30fps)。数据集总共有1个episode、2702帧和1个任务,使用so_follower机器人类型,数据以parquet和mp4格式存储,并分为训练集。
This dataset is created using LeRobot and is a robotics dataset focused on pick-and-place tasks. It includes action and observation data: the action data consists of 6 joint positions, such as shoulder translation, shoulder lift, elbow bend, wrist bend, wrist rotation, and gripper position; the observation data includes the same joint position states, along with front-view images and annotated images (stored in video format with a resolution of 480x640, 3 channels, and a frame rate of 30fps). The dataset contains a total of 1 episode, 2702 frames, and 1 task, uses the so_follower robot type, and the data is stored in parquet and mp4 formats, divided into a training set.
提供机构:
anikitakis创建时间:
2026-06-06
原始信息汇总
数据集概述:rollout_pick_n_place_dagger_r7
许可证: Apache-2.0
任务类别: 机器人学(Robotics)
标签: LeRobot
创建工具: 使用 LeRobot 创建
数据集结构
- 总集数 (Episodes): 1
- 总帧数 (Frames): 2702
- 总任务数 (Tasks): 1
- 帧率 (FPS): 30
- 数据文件大小 (Data Files): 100 MB
- 视频文件大小 (Video Files): 200 MB
- 分块大小 (Chunks Size): 1000
- 数据路径: data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet
- 视频路径: videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4
- 机器人类型: so_follower
- 数据划分: 仅包含训练集(train: "0:1")
特征(Features)
| 特征名称 | 数据类型 | 形状 | 说明 |
|---|---|---|---|
| action | float32 | [6] | 机械臂6个关节的动作:shoulder_pan, shoulder_lift, elbow_flex, wrist_flex, wrist_roll, gripper(位置) |
| observation.state | float32 | [6] | 机械臂6个关节的状态(与action名称一致) |
| observation.images.front | 视频 (video) | [480, 640, 3] | 前置摄像头RGB图像,分辨率480×640,3通道;编码格式av1,帧率30 FPS |
| observation.images.annotated | 视频 (video) | [480, 640, 3] | 标注后的前置摄像头RGB图像,分辨率480×640,3通道;编码格式av1,帧率30 FPS |
| timestamp | float32 | [1] | 时间戳 |
| frame_index | int64 | [1] | 帧索引 |
| episode_index | int64 | [1] | 集数索引 |
| index | int64 | [1] | 全局索引 |
| task_index | int64 | [1] | 任务索引 |
用途说明
该数据集适用于机器人操作任务中的模仿学习(Imitation Learning)或行为克隆(Behavior Cloning),特别针对“拾取与放置(Pick and Place)”任务,数据由DAgger(数据集聚合)方法采集。
可视化
可通过以下空间在线可视化数据集:



