rollout_R3search_30episodes_2cm_20260612_154521

该数据集使用LeRobot创建，属于机器人学领域，专门用于机器人控制和研究。数据集包含多个字段：动作（action）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等6个浮点值；观测状态（observation.state）与动作字段相同，表示机器人关节状态；观测图像包括顶部视角（observation.images.top）和腕部视角（observation.images.wrist），均为480x640分辨率的RGB视频，帧率为30fps，使用AV1编码；干预（intervention）为布尔值，表示是否有人为干预；时间戳（timestamp）为浮点值；帧索引（frame_index）、剧集索引（episode_index）、索引（index）和任务索引（task_index）均为整数值。数据集总共有5个剧集，2035帧，数据以Parquet格式存储，视频以MP4格式存储，总数据大小约为100MB，视频大小约为200MB。机器人类型为so_follower，适用于训练和测试机器人控制算法。This dataset is created using LeRobot and belongs to the field of robotics, specifically designed for robot control and research. It includes multiple fields: action consists of 6 floating-point values for shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position; observation.state is identical to the action field, representing the robot joint state; observation.images includes top view (observation.images.top) and wrist view (observation.images.wrist), both being RGB videos with a resolution of 480x640, a frame rate of 30fps, and encoded with AV1; intervention is a boolean value indicating whether human intervention occurred; timestamp is a floating-point value; frame_index, episode_index, index, and task_index are all integer values. The dataset contains a total of 5 episodes and 2035 frames, with data stored in Parquet format and videos in MP4 format. The total data size is approximately 100MB, and the video size is about 200MB. The robot type is so_follower, suitable for training and testing robot control algorithms.