mbodiai/ABB_PandG_20250328_152108

这个数据集包含了机器人操作相关的图像、指令、动作状态和奖励等信息。具体来说，它具有以下特征：观察部分包括图像和字符串类型的指令与提示；动作部分包括姿态（由x、y、z、roll、pitch、yaw六个浮点数表示）、抓握力度和机械臂末端执行器的力量；状态部分包括两个摄像头的图像和深度信息、世界中的物体以及机械臂末端执行器的姿态序列；奖励部分是一个浮点数，表示某一动作的奖励值；元数据部分包括序列索引和步索引。数据集分为训练集，提供了相应的字节数和示例数。

This dataset includes robot operation-related images, instructions, action states, and rewards. Specifically, it has the following features: the observation part includes images and string-type instructions and prompts; the action part includes pose (represented by six floats: x, y, z, roll, pitch, yaw), grasp force, and gripper force; the state part includes images and depth information from two cameras, objects in the world, and a sequence of gripper poses; the reward part is a float representing the reward for an action; the metadata part includes sequence index and step index. The dataset is split into a training set with provided byte size and example count.