videron/bottle-opening-combined-delta-stats

该数据集是一个用于机器人学习的数据集，专注于开瓶任务，由LeRobot工具创建。数据集包含170个情节，总计133,761帧，帧率为30fps，使用Apache 2.0许可证。数据来自anvil_openarm机器人，采集了多个视角的图像观察（包括腰部、胸部、左右手腕的摄像头，分辨率为480x640，RGB三通道），以及机器人的状态（16维关节位置）、动作（16维关节控制）、速度（16维关节速度）和力矩（16维关节力矩）信息。数据集结构以Parquet文件格式存储，适用于机器人控制和模仿学习任务。This dataset is a robotics learning dataset focused on bottle-opening tasks, created using the LeRobot tool. It contains 170 episodes with a total of 133,761 frames at 30fps, licensed under Apache 2.0. The data is collected from an anvil_openarm robot, featuring multi-view image observations (from waist, chest, left wrist, and right wrist cameras, with 480x640 resolution and RGB three channels), as well as robot state (16-dimensional joint positions), actions (16-dimensional joint controls), velocity (16-dimensional joint velocities), and effort (16-dimensional joint torques). The dataset is structured in Parquet file format and is suitable for robotics control and imitation learning tasks.