ZhaoRunyi/Piper_beaker_mixer_0421

Piper LeRobot数据集是一个用于机器人学习的数据集，基于ALOHA双机械臂遥操作数据生成。数据集包含左右机械臂的关节状态（各6个关节）、夹爪控制信号、末端执行器位置（3D坐标）和旋转（使用rot6d格式），以及来自高分辨率摄像头和左右腕部摄像头的图像观测。数据以10帧/秒（FPS）的速率采样，采用完整采样模式，并经过时间戳对齐处理。数据集适用于机器人控制、模仿学习、强化学习等任务，提供了丰富的多模态观测和动作数据。

The Piper LeRobot dataset is a dataset for robot learning, generated from teleoperation data of a dual-arm ALOHA robot. It includes joint states (6 joints per arm), gripper control signals, end-effector positions (3D coordinates), and rotations (in rot6d format) for both left and right arms, along with image observations from a high-resolution camera and left/right wrist cameras. Data is sampled at 10 FPS in full sampling mode, with timestamp alignment. The dataset is suitable for tasks such as robot control, imitation learning, and reinforcement learning, providing rich multimodal observation and action data.