ankile/real01c-insert-marker-d1-r2-stage1-candidate-screen-sobol10
收藏资源简介:
该数据集是一个用于机器人学习任务的数据集,由LeRobot工具创建。数据集包含60个总episodes和20556帧,使用Franka机器人收集。数据以Parquet格式存储,并包含多种特征:观测状态(如笛卡尔位置、关节位置、速度、夹爪位置)、动作(如笛卡尔速度、位置、关节速度、位置)、步骤数、干预标志、成功标志、奖励等。此外,数据集还包含两个摄像头视角的图像数据,分辨率为480x640,帧率为15fps,用于视觉输入。数据集适用于训练和评估机器人控制算法,如模仿学习或强化学习。
This dataset is designed for robotics learning tasks and was created using the LeRobot tool. It consists of 60 total episodes and 20556 frames, collected with a Franka robot. The data is stored in Parquet format and includes various features: observation states (such as Cartesian position, joint position, velocity, gripper position), actions (such as Cartesian velocity, position, joint velocity, position), steps to go, intervention flags, success flags, rewards, and more. Additionally, the dataset contains image data from two camera perspectives with a resolution of 480x640 and a frame rate of 15fps for visual input. It is suitable for training and evaluating robot control algorithms, such as imitation learning or reinforcement learning.




