mbodiai/ABB_PandG_20250402_124821

该数据集包含了观察、动作、状态、奖励和元数据等多个维度的信息。观察部分包含图像和文本指令，动作部分包含姿态（包括x、y、z、roll、pitch、yaw六个维度）、抓取力度和夹爪力度。状态部分包括两个相机视角的图像和深度信息、世界中的物体信息以及夹爪的姿态。奖励是一个浮点数值。数据集分为训练集，共有14个示例，总大小为148MB。此外，提供了默认配置，指定了训练集数据文件的路径。

The dataset includes information from multiple dimensions such as observation, action, state, reward, and metadata. The observation part contains images and text instructions, the action part includes pose (with six dimensions: x, y, z, roll, pitch, yaw), grasp force, and gripper force. The state part includes images and depth information from two camera perspectives, objects in the world, and the grippers pose. The reward is a floating-point number. The dataset is split into a training set with a total of 14 examples and a size of 148MB. Additionally, a default configuration is provided, specifying the path to the data files for the training set.