2026-07-24_shakeitup7_accelero_10kHz_nfft_64

该数据集是一个专为机器人学习设计的数据集，由LeRobot创建。它包含机器人的动作和状态观测数据：动作数据涵盖6个关节位置（具体包括肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部滚动和夹爪位置）；观测数据包括状态观测（同样为6个关节位置）和图像观测，图像观测由顶部摄像头视频（分辨率480x640像素，3通道，30fps）、腕部摄像头视频（分辨率480x640像素，3通道，30fps）以及触觉加速度计谱图视频（分辨率224x224像素，3通道，30fps）组成。数据集还包含元数据，如时间戳、帧索引和episode索引。总共有60个episodes，49437帧，数据以Parquet格式存储，总数据大小约为300MB（其中数据文件100MB，视频文件200MB）。机器人类型为so101_follower_dragon_tactile_bench，适用于训练任务。This dataset is designed for robot learning and created using LeRobot. It contains robot action and state observation data: action data includes 6 joint positions (specifically shoulder translation, shoulder lift, elbow bend, wrist bend, wrist roll, and gripper position); observation data includes state observations (also 6 joint positions) and image observations, which consist of a top camera video (resolution 480x640 pixels, 3 channels, 30fps), a wrist camera video (resolution 480x640 pixels, 3 channels, 30fps), and a tactile accelerometer spectrogram video (resolution 224x224 pixels, 3 channels, 30fps). The dataset also includes metadata such as timestamps, frame indices, and episode indices. In total, there are 60 episodes with 49437 frames, stored in Parquet format, with a total data size of approximately 300MB (100MB for data files and 200MB for video files). The robot type is so101_follower_dragon_tactile_bench, and it is suitable for training tasks.