notstrober/desk-stationery-to-drawer-bimanual_v4
收藏Hugging Face2026-05-14 更新2026-05-31 收录
下载链接:
https://hf-mirror.com/datasets/notstrober/desk-stationery-to-drawer-bimanual_v4
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是一个用于机器人操作的双机械臂数据集,专注于将桌面上的文具物品放入抽屉的任务。数据集包含91个完整操作序列(episodes),总计201,819帧数据,帧率为30fps。数据特征包括12维的动作和状态信息,分别对应左右机械臂的肩部、肘部、腕部和夹爪的12个关节位置。观测数据包括三个相机视角的RGB视频(左前、右前、右顶),分辨率为480x640。数据集采用Apache 2.0许可证,由LeRobot工具创建,适用于机器人学习和模仿学习研究。
This dataset is a bimanual robotic manipulation dataset focused on the task of placing desk stationery items into a drawer. It contains 91 complete episodes with a total of 201,819 frames at 30 fps. The features include 12-dimensional action and state information corresponding to 12 joint positions of the left and right robotic arms (shoulder, elbow, wrist, and gripper). Observations include RGB videos from three camera views (left front, right front, right top) at 480x640 resolution. The dataset is licensed under Apache 2.0, created using the LeRobot tool, and is suitable for robotics learning and imitation learning research.
提供机构:
notstrober


