eval_act_collect_empty_bottle_black_white_wrist_2k_bs8_k20

这是一个机器人技术数据集，使用LeRobot平台创建。数据集包含1个完整的情节，总计1334帧数据，采样率为30fps。数据特征包括9维的动作向量（控制机械臂的肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置，以及x、y和角速度），9维的状态观测向量（与动作相同的关节位置和速度），480x640分辨率的腕部摄像头视频数据，以及时间戳、帧索引、情节索引等元数据。数据集专为机器人学习和控制任务设计。

This is a robotics dataset created using the LeRobot platform. The dataset contains 1 full episode, with a total of 1334 frames of data sampled at 30fps. The data features include a 9-dimensional action vector (controlling the robotic arm's shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, gripper position, as well as x, y and angular velocities), a 9-dimensional state observation vector (containing the same joint positions and velocities as those in the action vector), 480x640 resolution wrist-mounted camera video data, as well as metadata such as timestamps, frame indices and episode indices. This dataset is specifically designed for robotic learning and control tasks.