red_to_brown_4
收藏资源简介:
该数据集是一个机器人操作数据集,使用LeRobot平台创建。数据集包含25个episodes,总计7497帧,采样频率为30fps。数据特征包括机器人的动作(6维关节位置:肩部平移、肩部抬升、肘部弯曲、腕部弯曲、腕部旋转、夹爪位置)和观测状态(相同的6维关节位置)。观测还包括两个摄像头的视频数据:顶部摄像头(cam_top)和腕部摄像头(cam_wrist),分辨率均为480x640,3通道彩色视频,编码格式为AV1。此外,数据集包含时间戳、帧索引、episode索引、任务索引等元数据。机器人类型为so_follower。数据以parquet文件格式存储,视频以MP4格式存储,总数据大小约为300MB(数据文件100MB,视频文件200MB)。数据集用于机器人学习任务,如模仿学习或强化学习。
This dataset is a robot manipulation dataset created using the LeRobot platform. It contains 25 episodes, totaling 7497 frames with a sampling frequency of 30fps. The data features include robot actions (6-dimensional joint positions: shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, gripper position) and observation states (the same 6-dimensional joint positions). Observations also include video data from two cameras: a top camera (cam_top) and a wrist camera (cam_wrist), both with a resolution of 480x640, 3-channel color video, encoded in AV1 format. Additionally, the dataset contains metadata such as timestamps, frame indices, episode indices, and task indices. The robot type is so_follower. Data is stored in parquet file format, with videos in MP4 format, and the total data size is approximately 300MB (100MB for data files and 200MB for video files). The dataset is intended for robot learning tasks, such as imitation learning or reinforcement learning.
数据集概述
- 名称:
red_to_brown_4 - 许可证: Apache-2.0
- 任务类别: 机器人学 (Robotics)
- 标签: LeRobot
数据集描述
该数据集使用 LeRobot 创建。
数据集结构
- 总片段数 (Episodes): 25
- 总帧数 (Frames): 7,497
- 总任务数 (Tasks): 1
- 帧率 (FPS): 30
- 数据格式: Parquet 文件 (
data/*/*.parquet) - 数据文件大小: 约 100 MB
- 视频文件大小: 约 200 MB
- 机器人类型:
so_follower
特征 (Features)
action:- 类型:
float32 - 形状:
[6] - 包含: 肩部俯仰、肩部升降、肘部弯曲、腕部弯曲、腕部旋转、夹爪位置
- 类型:
observation.state:- 类型:
float32 - 形状:
[6] - 包含: 与
action相同的6个机器人关节位置
- 类型:
observation.images.cam_top:- 类型:
video - 形状:
[480, 640, 3](高、宽、通道) - 编码: AV1, 30 FPS
- 类型:
observation.images.cam_wrist:- 类型:
video - 形状:
[480, 640, 3](高、宽、通道) - 编码: AV1, 30 FPS
- 类型:
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]
数据划分
- 训练集 (Train): 片段 0 至 24(共25个片段),即全部数据用于训练
其他信息
- 主页: 待补充
- 论文: 待补充
- 引用: 待补充



