pick_cube_franka_panda_10
收藏Hugging Face2026-05-26 更新2026-05-26 收录
下载链接:
https://huggingface.co/datasets/Lotussheep/pick_cube_franka_panda_10
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是一个机器人学数据集,使用LeRobot创建,专门用于机器人操作任务,特别是涉及机械臂(如Franka Panda)拾取立方体的场景。数据集包含10个总集数,233个总帧数,和1个总任务数。数据结构包括动作特征(如delta_x、delta_y、delta_z和gripper控制)、奖励和完成状态、观察图像(前部和腕部摄像头视频,分辨率为128x128,帧率为10fps,使用AV1编解码器)、机器人状态(18维浮点数)、时间戳、帧索引、集索引、任务索引等。数据以Parquet格式存储,视频以MP4格式存储,总数据大小约100MB,视频大小约200MB。数据集适用于机器人强化学习或模仿学习任务,并已分割为训练集(集0到10)。
This is a robotics dataset developed with LeRobot, specifically tailored for robotic manipulation tasks, especially scenarios where robotic arms (e.g., Franka Panda) pick up cubes. The dataset contains 10 total episodes, 233 total frames, and 1 single task. Its data structure encompasses action features (e.g., delta_x, delta_y, delta_z, and gripper control), rewards and completion statuses, observation videos from front and wrist cameras (with a resolution of 128×128, 10fps frame rate, and AV1 codec), 18-dimensional floating-point robot states, timestamps, frame indices, episode indices, and task indices, among others. The tabular data is stored in Parquet format, while the video files are stored in MP4 format. The total size of the tabular data is approximately 100 MB, and the total size of the videos is around 200 MB. This dataset is applicable to robotic reinforcement learning or imitation learning tasks, and has been split into a training set covering episodes 0 to 10.
提供机构:
Lotussheep创建时间:
2026-05-26
原始信息汇总
数据集名称
pick_cube_franka_panda_10
许可协议
Apache-2.0
任务类别
机器人学(robotics)
标签
LeRobot
创建工具
该数据集使用 LeRobot 创建。
数据集结构
版本:v3.0
帧率:10 FPS
总样本量:
- 总片段数:10
- 总帧数:233
- 总任务数:1
数据划分:所有数据均用于训练(splits.train = "0:10")
文件大小:数据文件约 100 MB,视频文件约 200 MB
数据存储路径:
- 数据文件:
data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet - 视频文件:
videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4
特征字段:
| 字段名 | 数据类型 | 形状 | 说明 |
|---|---|---|---|
action |
float32 | [4] | 动作:delta_x, delta_y, delta_z, gripper |
next.reward |
float32 | [1] | 奖励信号 |
next.done |
bool | [1] | 是否结束 |
complementary_info.discrete_penalty |
float32 | [1] | 离散惩罚项 |
observation.images.front |
video | [3, 128, 128] | 前方摄像头图像(128x128 RGB) |
observation.images.wrist |
video | [3, 128, 128] | 腕部摄像头图像(128x128 RGB) |
observation.state |
float32 | [18] | 机器人状态(18维) |
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 |
视频编码:使用 AV1 编码,分辨率 128x128,帧率 10 FPS,每个摄像头含 3 个通道。



