oxe-aug/fractal20220817_data_train_10000_15000_augmented

fractal20220817_data_train_10000_15000_augmented是一个基于OXE-Aug的增强机器人数据集，包含5000个剧集，215178帧图像，涵盖8种不同类型的机器人，如jaco、kinova3、kuka_iiwa等。数据集以3帧每秒的速率记录，每个视频包含剧集索引、帧索引、时间戳等信息，以及对应机器人的图像、末端执行器位置、关节角度等数据。数据集适用于机器人控制和交叉Embodiment策略学习等研究。

The fractal20220817_data_train_10000_15000_augmented is an augmented robot dataset based on OXE-Aug, containing 5,000 episodes, 215,178 frames of images, covering 8 different types of robots such as jaco, kinova3, kuka_iiwa, etc. The dataset is recorded at a rate of 3 frames per second, with each video including episode index, frame index, timestamp, and other information, as well as corresponding robot images, end-effector positions, joint angles, and other data. The dataset is suitable for research on robot control and cross-Embodiment policy learning.