mbodiai/ABB_PandG_20250331_214316

该数据集包含了观察、动作、状态、奖励和元数据等信息，观察中包括图片、指令和提示信息，动作包括姿态、抓取力度和机械臂力量等，状态包括双摄像头图片、深度信息、世界中的物体和机械臂姿态等，奖励为浮点数值，元数据包括剧集索引和步骤索引。数据集分为训练集，共有9个样本。

The dataset includes information such as observations, actions, states, rewards, and metadata. Observations consist of images, instructions, and prompt messages. Actions include posture, grasping force, and robotic arm force. States include dual camera images, depth information, objects in the world, and robotic arm posture. Rewards are represented by floating-point numbers, and metadata includes episode index and step index. The dataset is split into a training set with a total of 9 samples.