mbodiai/ABB_PandG_20250401_155126

这个数据集包含机器人的观察数据、行为数据、状态数据、奖励值以及元数据。观察数据中包括图像、指令和提示信息。行为数据包括机器人的姿态、抓握力度和夹爪力度。状态数据包括两个摄像头视角的图像、深度信息、世界中的物体信息和夹爪姿态。奖励值是浮点数类型，表示某一步动作的奖励。元数据包括剧集索引和步骤索引。数据集分为训练集，共有6个示例。

The dataset includes robot observation data, action data, state data, reward values, and metadata. Observation data contain images, instructions, and prompts. Action data include robot poses, grasp strength, and gripper force. State data include images from two camera views, depth information, objects in the world, and gripper pose. The reward value is a float type, representing the reward for a certain steps action. Metadata include episode index and step index. The dataset is split into a training set with a total of 6 examples.