taccap-g1-Peel-arrange-close-earphones-Velcro-0728
收藏资源简介:
该数据集是使用LeRobot工具创建的机器人学数据集,专门针对双TacCap夹持器机器人(bi_taccap_gripper)。它包含130个episodes,总计416,775帧,帧率为30fps,数据文件大小为100MB,视频文件大小为200MB。数据特征包括动作(20维浮点数组,表示左右夹持器的TCP位置、旋转和夹持器位置)、观测状态(与动作相同维度)、以及多个触觉图像(左右触觉左右传感器,分辨率400x700)和手腕图像(左右手腕,分辨率480x640),所有图像均为视频格式(H.264编码,YUV420p像素格式)。此外,还包含时间戳、帧索引、episode索引、任务索引等元数据。数据集用于训练和评估机器人控制算法,支持触觉感知和视觉反馈。
This dataset is a robotics dataset developed using the LeRobot tool, tailored specifically for the bi_taccap_gripper (dual TacCap gripper) robot. It contains 130 episodes, totaling 416,775 frames at a frame rate of 30 fps. The total size of the data files is 100 MB, while the video files amount to 200 MB. The data features include actions (a 20-dimensional floating-point array representing the TCP position, rotation, and gripper position of both left and right grippers), observation states (with the same dimensionality as the actions), multiple tactile images from the left and right tactile sensors with a resolution of 400×700, and wrist images from the left and right wrists with a resolution of 480×640. All images are in video format with H.264 encoding and YUV420p pixel format. In addition, the dataset includes metadata such as timestamps, frame indices, episode indices, and task indices. This dataset is intended for training and evaluating robot control algorithms, and supports tactile perception and visual feedback.
数据集概述
- 数据集名称:taccap-g1-Peel-arrange-close-earphones-Velcro-0728
- 机器人类型:bi_taccap_gripper(双臂触觉夹爪)
- 总片段数:130
- 总帧数:416775
- 总任务数:1(剥取、整理、闭合耳机的魔术贴)
- 帧率:30 FPS
- 许可协议:Apache-2.0
- 创建工具:LeRobot
数据划分
- 训练集:全部130个片段(索引0至129)
数据存储
- 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(H.264编码),路径为
videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4 - 数据文件大小:约100 MB
- 视频文件大小:约200 MB
- 分块大小:每块1000帧
特征说明
动作与状态(action 和 observation.state)
- 数据类型:float32
- 形状:[20]
- 含义:包含左右机械臂的TCP位姿(位置x,y,z + 旋转量r1~r6)和夹爪开合状态,具体顺序为:
left_tcp.x,left_tcp.y,left_tcp.z,left_tcp.r1,left_tcp.r2,left_tcp.r3,left_tcp.r4,left_tcp.r5,left_tcp.r6,left_gripper.posright_tcp.x,right_tcp.y,right_tcp.z,right_tcp.r1,right_tcp.r2,right_tcp.r3,right_tcp.r4,right_tcp.r5,right_tcp.r6,right_gripper.pos
视觉观察(observation.images)
- 左触觉左视图 (
left_tactile_left):视频,400x700x3,H.264,30 FPS - 左触觉右视图 (
left_tactile_right):视频,400x700x3,H.264,30 FPS - 左腕部视图 (
left_wrist):视频,480x640x3,H.264,30 FPS - 右触觉左视图 (
right_tactile_left):视频,400x700x3,H.264,30 FPS - 右触觉右视图 (
right_tactile_right):视频,400x700x3,H.264,30 FPS - 右腕部视图 (
right_wrist):视频,480x640x3,H.264,30 FPS
其他元数据
timestamp:float32,形状[1],时间戳frame_index:int64,形状[1],帧索引episode_index:int64,形状[1],片段索引index:int64,形状[1],全局索引task_index:int64,形状[1],任务索引
引用信息
bibtex @misc{xense-taccap-lerobot, author = {XenseRobotics Team}, title = {LeRobot-Xense: LeRobot with Xense Tactile Robotics Support}, howpublished = {url{https://github.com/Vertax42/xense-taccap-lerobot}}, year = {2026} }



