cell1_20260516_youssef-mohamed_9pm-12pm_lego20260517_001325_upload
收藏Hugging Face2026-05-17 更新2026-05-17 收录
下载链接:
https://huggingface.co/datasets/nodogoro/cell1_20260516_youssef-mohamed_9pm-12pm_lego20260517_001325_upload
下载链接
链接失效反馈官方服务:
资源简介:
这是一个机器人学数据集,使用LeRobot工具创建,专注于机器人操作任务。数据集包含2个完整任务序列(episodes),总计10587帧数据,涉及1种任务类型。数据采用parquet格式存储,包含动作、状态观测和视觉观测等多模态信息。具体特征包括:14维动作向量(两个机械臂的6自由度位姿和夹爪宽度)、32维状态向量(编码器角度、IMU加速度/角速度、机械臂位姿、夹爪开合状态等)、以及来自多个摄像头的RGB和深度图像(包括左右腕部摄像头和两个基座摄像头,分辨率分别为480x640和768x1024,30fps视频格式)。此外,还包含时间戳、帧索引、任务索引等元数据。机器人平台为starpilot_yam_gripper。数据集适用于机器人模仿学习、强化学习等研究。
This is a robotics dataset created using the LeRobot tool, focusing on robot manipulation tasks. The dataset contains 2 complete task sequences (episodes), totaling 10587 frames of data, involving 1 task type. The data is stored in parquet format and includes multimodal information such as actions, state observations, and visual observations. Specific features include: a 14-dimensional action vector (6-degree-of-freedom poses for two robotic arms and gripper width), a 32-dimensional state vector (encoder angles, IMU acceleration/angular velocity, robotic arm poses, gripper open/close status, etc.), and RGB and depth images from multiple cameras (including left and right wrist cameras and two base cameras, with resolutions of 480x640 and 768x1024, in 30fps video format). Additionally, it includes metadata such as timestamps, frame indices, and task indices. The robot platform is starpilot_yam_gripper. The dataset is suitable for research in robot imitation learning, reinforcement learning, and related fields.
提供机构:
nodogoro
创建时间:
2026-05-17
原始信息汇总
数据集概述
- 数据集名称: cell1_20260516_youssef-mohamed_9pm-12pm_lego20260517_001325_upload
- 许可证: Apache-2.0
- 任务类别: 机器人学 (Robotics)
- 创建工具: LeRobot
数据集规模与结构
- 总片段数 (Episodes): 2
- 总帧数 (Frames): 10,587
- 总任务数 (Tasks): 1
- 帧率 (FPS): 30
- 数据文件大小: 约 100 MB
- 视频文件大小: 约 200 MB
- 数据切块大小 (Chunk Size): 1000
- 划分: 所有数据用于训练 (train: "0:2")
机器人类型
- 机器人: starpilot_yam_gripper(双机械臂夹爪机器人)
特征空间
| 特征名称 | 数据类型 | 形状 | 描述 |
|---|---|---|---|
| action | float32 | (14,) | 包含两个机械臂的关节位置与夹爪宽度(arm1/arm2 各7维:x, y, z, rx, ry, rz, grip_width) |
| observation.state | float32 | (32,) | 包含两个机械臂的编码器读数、IMU数据、位姿(位置+四元数)、夹爪开合状态及距离(arm1/arm2 各16维) |
| observation.images.left_wrist_0_camera_rgb_image | video (h264) | (480, 640, 3) | 左腕相机RGB图像,30fps |
| observation.images.left_wrist_0_camera_rgb_image_depth | video (h264) | (480, 640, 3) | 左腕相机深度图像(存储为3通道),30fps |
| observation.images.right_wrist_0_camera_rgb_image | video (h264) | (480, 640, 3) | 右腕相机RGB图像,30fps |
| observation.images.right_wrist_0_camera_rgb_image_depth | video (h264) | (480, 640, 3) | 右腕相机深度图像(存储为3通道),30fps |
| observation.images.base_0_camera_rgb_image | video (h264) | (480, 640, 3) | 基座0号相机RGB图像,30fps |
| observation.images.base_1_camera_rgb_image | video (h264) | (768, 1024, 3) | 基座1号相机RGB图像(更高分辨率),30fps |
| 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 |
数据存储格式
- 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 - 元数据文件:
meta/info.json



