libero_90_iiwa_defaultCams
收藏资源简介：
该数据集使用LeRobot工具创建，专注于机器人技术领域，具体涉及libero和iiwa机器人平台。数据集包含3256个episodes，总计472632帧，覆盖67个不同任务，以支持机器人学习和控制研究。数据以parquet文件和mp4视频格式存储，总数据文件大小为100MB，视频文件大小为200MB，帧率为20fps，便于高效访问和处理。数据集仅包含训练集，适用于模型训练和验证。特征包括：观测状态，如8维状态向量、6维末端执行器状态、7维关节状态、2维夹爪状态；图像观测，提供256x256 RGB图像，包括主图像和腕部图像，以捕捉视觉信息；7维动作向量，表示机器人控制指令；以及时间戳、帧索引、episode索引、任务索引等元数据，用于数据组织和分析。机器人类型为iiwa，适用于机器人控制、模仿学习等任务，旨在促进机器人智能算法的开发和应用。
This dataset was created using the LeRobot toolkit, focusing on the field of robotics, specifically involving the LIBERO and iiwa robotic platforms. It contains 3256 episodes, totaling 472,632 frames, covering 67 distinct tasks to support robotics learning and control research. The data is stored in Parquet files and MP4 video formats, with a total data file size of 100 MB and video file size of 200 MB, and a frame rate of 20 fps, enabling efficient access and processing. This dataset only includes a training split, suitable for model training and validation. The included features are as follows: observational states such as an 8-dimensional state vector, 6-dimensional end-effector state, 7-dimensional joint state, and 2-dimensional gripper state; image observations providing 256×256 RGB images including main and wrist-mounted images to capture visual information; a 7-dimensional action vector representing robotic control commands; as well as metadata such as timestamps, frame indices, episode indices, and task indices for data organization and analysis. The robotic platform used is the iiwa, applicable to tasks such as robotic control and imitation learning, aiming to facilitate the development and application of robotic intelligent algorithms.
数据集概述
数据集名称：libero_90_iiwa_defaultCams
数据集地址：https://huggingface.co/datasets/OliverHausdoerfer/libero_90_iiwa_defaultCams
许可证：Apache-2.0
任务类别：机器人学 (Robotics)
创建工具：基于 LeRobot 框架创建
数据集规模
- 总片段数 (Episodes)：3256
- 总帧数 (Frames)：472,632
- 总任务数 (Tasks)：67
- 数据文件大小：100 MB
- 视频文件大小：200 MB
- 帧率 (FPS)：20
- 数据集划分：仅包含训练集 (train: 0:3256)
机器人类型
- 机器人：KUKA iiwa
数据格式
- 数据文件：Parquet 格式，路径为
data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet - 视频文件：MP4 格式，路径为
videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4 - 数据分块大小：每块 1000 个样本
特征 (Features)
| 特征名称 | 数据类型 | 形状 | 说明 |
|---|---|---|---|
observation.state |
float32 | (8,) | 机器人状态：x, y, z, axis_angle1-3, gripper (2维) |
observation.states.ee_state |
float32 | (6,) | 末端执行器状态：x, y, z, axis_angle1-3 |
observation.states.joint_state |
float32 | (7,) | 关节状态：joint_0 到 joint_6 |
observation.states.gripper_state |
float32 | (2,) | 夹爪状态：gripper (2维) |
observation.images.image |
video | (256, 256, 3) | 主摄像头图像，分辨率 256x256，RGB，AV1 编码，20 FPS |
observation.images.wrist_image |
video | (256, 256, 3) | 腕部摄像头图像，分辨率 256x256，RGB，AV1 编码，20 FPS |
action |
float32 | (7,) | 动作指令：x, y, z, axis_angle1-3, gripper |
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 |
视频编码信息
- 编解码器：AV1
- 像素格式：yuv420p
- 分辨率：256 x 256
- 帧率：20 FPS
- 通道数：3 (RGB)
- 无深度图：否 (is_depth_map: false)
- 无音频：是 (has_audio: false)
引用信息
- 主页：[More Information Needed]
- 论文：[More Information Needed]
- BibTeX：[More Information Needed]
