02062026_newtest
收藏资源简介:
该数据集是一个机器人控制数据集,使用LeRobot工具创建,针对Agilex Piper双手机器人。它包含3个完整任务片段,总计1344帧数据,帧率为20fps。数据集采用Apache 2.0许可证。数据以parquet格式存储,视频以mp4格式存储。数据集包含多模态观测数据:机器人状态(14维浮点数,对应左右机械臂各6个关节角度和左右夹爪状态)、动作数据(相同结构)、三个视角的RGB图像(前、右、左,分辨率640x480)以及对应的时间戳和相机时间戳。此外,还包括帧索引、片段索引和任务索引等元数据。该数据集适用于机器人学习、模仿学习或强化学习任务,特别是双手机器人控制场景。
This is a robotic control dataset created with the LeRobot toolkit, targeting the Agilex Piper dual-arm robot. It includes 3 complete task segments, totaling 1344 frames of data at a frame rate of 20fps. The dataset is licensed under Apache 2.0. Structured dataset entries are stored in Parquet format, while video files are saved in MP4 format. The dataset contains multimodal observational data: robot state (14-dimensional floating-point values, corresponding to 6 joint angles and gripper status for each of the left and right robotic arms), action data with the same structure, RGB images from three viewpoints (front, right, left) with a resolution of 640×480, as well as corresponding timestamps and camera-specific timestamps. Additionally, it includes metadata such as frame indices, segment indices, and task indices. This dataset is suitable for robotics learning, imitation learning, or reinforcement learning tasks, especially dual-arm robot control scenarios.
- 许可协议: Apache-2.0
- 任务类别: 机器人学 (Robotics)
- 标签: LeRobot
- 数据集大小:
- 总片段数 (Episodes): 3
- 总帧数 (Frames): 1344
- 总任务数 (Tasks): 1
- 数据文件大小: 100 MB
- 视频文件大小: 200 MB
- 帧率 (FPS): 20
- 机器人类型: agilex_piper_bimanual (双机械臂)
- 数据分割: 训练集 (train) 包含全部3个片段
- 数据格式: Parquet (数据) + MP4 (视频)
- 特征 (Features):
- 观测状态 (observation.state): 14维浮点向量,包含左右各6个关节角度和2个夹爪状态
- 动作 (action): 14维浮点向量,与观测状态结构相同
- 观测图像:
- 正面 (observation.images.front): 640x480 像素,3通道,H.264编码,20 FPS
- 右侧 (observation.images.right): 640x480 像素,3通道,H.264编码,20 FPS
- 左侧 (observation.images.left): 640x480 像素,3通道,H.264编码,20 FPS
- 相机时间戳:
observation.camera_ts_ns.front.rgb: 整数时间戳observation.camera_ts_ns.right.rgb: 整数时间戳observation.camera_ts_ns.left.rgb: 整数时间戳
- 遥测数据:
observation.telemetry.right.joint_angles_deg: 6维浮点,右臂关节角度(度)observation.telemetry.left.joint_angles_deg: 6维浮点,左臂关节角度(度)
- 元数据:
timestamp(浮点),frame_index(整数),episode_index(整数),index(整数),task_index(整数)



