learner1119/ffw_sh5_rev1_pick_white_cup
收藏资源简介:
该数据集使用Physical AI Tools和LeRobot工具创建,专注于机器人学领域,特别是针对ffw_sh5_rev1型号机器人。数据集包含5个训练集,总帧数为1927,帧率为10fps。数据以parquet文件格式存储,并包含视频文件。数据集结构包括时间戳、帧索引、集索引、任务索引等元数据,以及观测数据(如头部摄像头图像和机器人状态)和动作数据。观测图像分辨率为1080x1920,观测状态和动作均包含54个维度,分别对应左右手臂和手指的关节角度。数据集旨在支持机器人控制和学习任务,适用于训练和评估机器人模型。
This dataset was created using Physical AI Tools and LeRobot, focusing on robotics tasks specifically for the ffw_sh5_rev1 robot model. It contains 5 training episodes with a total of 1927 frames at a frame rate of 10fps. The data is stored in parquet format and includes video files. The dataset structure comprises metadata such as timestamp, frame index, episode index, task index, as well as observation data (e.g., head camera images and robot state) and action data. Observation images have a resolution of 1080x1920, while observation state and action each include 54 dimensions corresponding to joint angles of left and right arms and fingers. The dataset is designed to support robot control and learning tasks, suitable for training and evaluating robot models.




