ecappiell/light_box_experiment_20260527_130455

该数据集是一个机器人学数据集，使用LeRobot工具创建，专注于机器人动作和状态观测。数据集包含动作数据（6个浮点数，表示肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪的位置）、状态观测数据（123个浮点数，涵盖关节位置、原始电流、负载、硬件速度、速度、原始扭矩、签名电流和扭矩、滤波速度、估计外部扭矩和方差、模型扭矩（如电机、重力、摩擦、科里奥利、惯性、背隙和模型扭矩）以及TCP力和力矩、HPI夹爪数据等）、图像观测数据（顶部和角落视角的视频，分辨率480x640，3通道，30fps，AV1编码）、时间戳、帧索引、episode索引、索引和任务索引。数据集结构包括1个episode、647帧、1个任务，数据以parquet文件（100MB）和视频文件（200MB）存储，机器人类型为so_follower，适用于机器人控制、模仿学习或强化学习任务。This dataset is a robotics dataset created using the LeRobot tool, focusing on robot actions and state observations. It includes action data (6 float32 values representing positions of shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper), state observation data (123 float32 values covering joint positions, raw currents, loads, hardware velocities, velocities, raw torques, signed currents and torques, filtered velocities, estimated external torques and variances, model torques such as motor, gravity, friction, Coriolis, inertial, backlash, and model torques, as well as TCP forces and torques, HPI gripper data, etc.), image observation data (videos from top and corner views with 480x640 resolution, 3 channels, 30fps, AV1 codec), timestamps, frame indices, episode indices, indices, and task indices. The dataset structure comprises 1 episode, 647 frames, 1 task, stored in parquet files (100MB) and video files (200MB), with a robot type of so_follower, suitable for robot control, imitation learning, or reinforcement learning tasks.