pick_up_bottle_one_45
收藏资源简介:
该数据集是一个机器人学数据集,由LeRobot创建,专注于双手机器人bi_so100_follower的控制任务。数据集包含45个episodes,总计5757帧数据,每帧包括动作特征(12个浮点数表示左右手臂各关节的位置)、观察状态(12个浮点数表示关节位置),以及来自三个摄像头的视频观察:头摄像头、左摄像头和右摄像头,视频分辨率为480x640,帧率为30fps,编码为av1。其他元数据包括时间戳、帧索引、episode索引、任务索引等。数据集用于训练和评估机器人控制算法,支持动作和视觉观察的联合分析。
This dataset is a robotics dataset created by LeRobot, focusing on the control task of the bimanual robot bi_so100_follower. It contains 45 episodes, totaling 5757 frames of data. Each frame includes action features (12 floating-point numbers representing the joint positions of the left and right arms), observation states (12 floating-point numbers representing joint positions), and video observations from three cameras: head camera, left camera, and right camera, with a resolution of 480x640, a frame rate of 30fps, and encoding in av1. Other metadata includes timestamps, frame indices, episode indices, task indices, etc. The dataset is used for training and evaluating robot control algorithms, supporting joint analysis of actions and visual observations.
数据集概述
- 数据集名称:
pick_up_bottle_one_45 - 数据集地址: https://huggingface.co/datasets/usrname1/pick_up_bottle_one_45
- 许可证: Apache-2.0
- 任务类别: 机器人学 (Robotics)
- 标签: LeRobot
数据集结构
- 代码库版本: v3.0
- 机器人类型: bi_so100_follower
- 总片段数: 45
- 总帧数: 5757
- 总任务数: 1
- 块大小: 1000
- 数据文件大小: 100 MB
- 视频文件大小: 500 MB
- 帧率: 30 FPS
- 数据集划分: 全部45个片段用于训练 (train: 0:45)
数据特征
动作 (action)
- 数据类型: float32
- 形状: [12]
- 字段: 左右肩膀、肘部、腕部、腕部旋转和夹爪的12个位置维度
观测状态 (observation.state)
- 数据类型: float32
- 形状: [12]
- 字段: 与动作相同的12个位置维度
观测图像 (observation.images)
- 头摄像头 (head_cam): 480x640 像素,3 通道,30 FPS,AV1 编码
- 左摄像头 (left_cam): 480x640 像素,3 通道,30 FPS,AV1 编码
- 右摄像头 (right_cam): 480x640 像素,3 通道,30 FPS,AV1 编码
其他特征
- timestamp: float32,形状 [1]
- frame_index: int64,形状 [1]
- episode_index: int64,形状 [1]
- index: int64,形状 [1]
- task_index: int64,形状 [1]
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
引用信息
- 引用: 待补充 (More Information Needed)



