KS325/open-upper-drawer-r2-occ-val

该数据集是通过LeRobot平台创建的机器人控制数据集，专门用于机器人学习任务。数据集包含4个完整的情节（episodes），总计2999帧数据，帧率为30fps。数据来源于so_follower类型的机器人，记录了机器人的动作（包括肩部平移、肩部升降、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置共6个关节的位置数据）和观测状态（同样为6维关节位置）。此外，数据集提供了两个摄像头的视觉观测数据（camera1和camera2），均为480x640分辨率的RGB视频流。数据集还包含时间戳、帧索引、情节索引、任务索引等元数据，以支持机器人强化学习或模仿学习的研究。数据以parquet格式存储，视频以mp4格式存储，遵循Apache 2.0许可证。

This dataset was created using the LeRobot platform and is designed for robotics learning tasks. It consists of 4 total episodes with 2999 frames at 30 fps. The data comes from a so_follower robot type and includes actions (6-dimensional joint positions for shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper) and observation states (also 6-dimensional joint positions). Additionally, the dataset provides visual observations from two cameras (camera1 and camera2), both as 480x640 resolution RGB video streams. It also contains metadata such as timestamps, frame indices, episode indices, and task indices to support research in robot reinforcement learning or imitation learning. The data is stored in parquet format, with videos in mp4 format, and is licensed under Apache 2.0.