so101_pickplace_aggregated

该数据集是一个机器人学数据集，专注于抓取和放置任务（pickplace），使用LeRobot工具创建。数据集包含48个回合（episodes）、31,291帧数据，涉及2个任务。数据特征包括动作数据（action），涵盖6个关节位置（如肩部平移、肩部升降、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置）；观测状态（observation.state），同样包含6个关节位置；以及来自腕部和前部摄像头的图像观测（observation.images.wrist和observation.images.front），图像分辨率为480x640，3通道，帧率为30fps，视频编码为av1。此外，数据集还包含时间戳、帧索引、回合索引、索引和任务索引等元数据。数据以Parquet文件格式存储，总数据文件大小为100MB，视频文件大小为200MB。数据集适用于机器人控制、强化学习或计算机视觉任务，特别是针对机械臂的抓取和放置操作。This dataset is a robotics dataset focused on pick-and-place tasks (pickplace), created using the LeRobot tool. It contains 48 episodes, 31,291 frames of data, and involves 2 tasks. Data features include action data, covering 6 joint positions (such as shoulder translation, shoulder lift, elbow flexion, wrist flexion, wrist rotation, and gripper position); observation states (observation.state), also including 6 joint positions; and image observations from wrist and front cameras (observation.images.wrist and observation.images.front), with image resolution of 480x640, 3 channels, frame rate of 30fps, and video encoding as av1. Additionally, the dataset includes metadata such as timestamps, frame indices, episode indices, indices, and task indices. Data is stored in Parquet file format, with a total data file size of 100MB and video file size of 200MB. The dataset is suitable for robotics control, reinforcement learning, or computer vision tasks, particularly for robotic arm pick-and-place operations.