Lisette1231/20260426_pickeggtoplate2

该数据集是一个机器人控制数据集，由LeRobot创建，专为机器人学习任务设计。数据集包含10个episodes，总计6171帧数据，使用Seeed B601 DM Follower机器人。数据以parquet文件和视频文件形式存储，视频帧率为30fps。数据集的特征包括：动作（7维浮点数，表示机器人关节位置）、状态观测（7维浮点数，同样表示关节位置）、来自腕部和前部摄像头的图像观测（分辨率480x640，3通道RGB视频），以及补充信息如策略动作、干预状态和时间戳等。数据集主要用于训练和评估机器人控制算法，支持模仿学习或强化学习任务。

This dataset is a robotics control dataset created by LeRobot, designed for robot learning tasks. It contains 10 episodes with a total of 6171 frames, using the Seeed B601 DM Follower robot. The data is stored in parquet files and video files, with a video frame rate of 30fps. The dataset features include: action (7-dimensional float32 representing robot joint positions), state observation (7-dimensional float32 also representing joint positions), image observations from wrist and front cameras (resolution 480x640, 3-channel RGB video), and complementary information such as policy action, intervention state, and timestamps. The dataset is primarily used for training and evaluating robot control algorithms, supporting tasks like imitation learning or reinforcement learning.