oxe-auge/fractal20220817_data_train_72500_75000_augmented

fractal20220817_data_train_72500_75000_augmented数据集是OXE-AugE项目的一部分，专注于机器人的增强数据。该数据集包含多种机器人的图像、关节位置和末端执行器姿态，以及自然语言指令。数据集旨在用于机器人跨体现策略的训练。具体包括2500个片段，109172帧，帧率为3fps，支持多种机器人型号如jaco、kinova3、kuka_iiwa、panda、sawyer、ur5e和xarm7。数据以parquet文件格式存储，包含丰富的特征如机器人图像、关节角度、末端执行器误差等。

The fractal20220817_data_train_72500_75000_augmented dataset is part of the OXE-AugE project, focusing on augmented data for robotics. It includes images, joint positions, and end-effector poses for various robots, along with natural language instructions. The dataset is designed for training cross-embodiment policies in robotics. It consists of 2500 episodes, 109172 frames, at 3fps, and supports multiple robot models such as jaco, kinova3, kuka_iiwa, panda, sawyer, ur5e, and xarm7. The data is stored in parquet files and includes rich features like robot images, joint angles, end-effector errors, etc.