pick_lego_block

该数据集是通过LeRobot创建的，涉及机器人技术，具体使用Franka机器人。数据集包含307帧数据，1个任务和1个片段。数据包括来自右腕摄像头和第三人称摄像头的图像观察（256x256x3），机器人状态观察（7维浮点数，包括位置、姿态和夹持器状态），以及动作数据（7维浮点数，同样包括位置、姿态和夹持器状态）。数据以parquet格式存储，帧率为30fps。

This dataset was developed using LeRobot, targeting robotics applications with the Franka robot. It consists of 307 frames of data, one task, and one segment. The data includes image observations (256x256x3) captured by both the right-wrist camera and a third-person camera, 7-dimensional floating-point robot state observations including position, orientation, and gripper status, as well as 7-dimensional floating-point action data that also covers position, orientation, and gripper status. The dataset is stored in Parquet format with a frame rate of 30 fps.

原始信息汇总

数据集概述

基本信息

• 名称: pick_lego_block
• 许可证: Apache-2.0
• 任务类别: 机器人学 (robotics)
• 标签: LeRobot
• 代码库版本: v2.1
• 机器人类型: franka

数据集结构

• 总集数: 1
• 总帧数: 307
• 总任务数: 1
• 总视频数: 0
• 总块数: 1
• 块大小: 1000
• 帧率: 30 fps
• 数据路径: data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet
• 视频路径: 无

特征

• observation.images.right_wrist_camera:
  • 类型: 图像
  • 形状: [256, 256, 3]
  • 维度名称: ["height", "width", "channels"]
• observation.images.right_third_person_camera:
  • 类型: 图像
  • 形状: [256, 256, 3]
  • 维度名称: ["height", "width", "channels"]
• observation.state:
  • 类型: float32
  • 形状: [7]
  • 维度名称: ["x", "y", "z", "roll", "pitch", "yaw", "gripper"]
• action:
  • 类型: float32
  • 形状: [7]
  • 维度名称: ["x", "y", "z", "roll", "pitch", "yaw", "gripper"]
• timestamp:
  • 类型: float32
  • 形状: [1]
• frame_index:
  • 类型: int64
  • 形状: [1]
• episode_index:
  • 类型: int64
  • 形状: [1]
• index:
  • 类型: int64
  • 形状: [1]
• task_index:
  • 类型: int64
  • 形状: [1]

引用

• BibTeX: [More Information Needed]