YOLO2431/combined_clean_up_table_v2

这是一个用于机器人学习的数据集，专注于双臂机器人（型号为yam_bimanual）的操控任务。数据集包含603个episodes、595,150帧和1,809个视频，覆盖7个不同的任务，所有数据均用于训练。数据以30fps的帧率采集，包括多模态特征：机器人的状态观察（如左右机械臂的位置、旋转6D表示、夹爪状态和关节角度）、动作命令，以及来自三个视角（头部、左手腕和右手腕）的图像视频（分辨率360x640，RGB格式）。此外，还提供时间戳、帧索引、episode索引和任务索引等元数据。该数据集适用于机器人控制、模仿学习或强化学习的研究和应用。This is a dataset for robotics learning, focusing on bimanual robot (model yam_bimanual) manipulation tasks. It contains 603 episodes, 595,150 frames, and 1,809 videos, covering 7 distinct tasks, all designated for training. Data is collected at 30fps and includes multimodal features: robot state observations (such as left and right arm positions, rotation 6D representations, gripper states, and joint angles), action commands, and image videos from three perspectives (head, left wrist, and right wrist) with a resolution of 360x640 in RGB format. Additionally, metadata such as timestamps, frame indices, episode indices, and task indices are provided. The dataset is suitable for research and applications in robot control, imitation learning, or reinforcement learning.