Alkatt/eval_LAVLA_Async_VI_100M_40k_b20_whiteblackcube_v2_2026_05_27_white_latency_bench_v5
收藏资源简介:
该数据集是一个机器人控制数据集,使用LeRobot工具创建。数据集针对so_follower机器人类型,包含2个训练集,总计933帧,数据以30fps的视频帧率记录。数据集结构包括动作数据(如肩部、肘部、腕部和抓取器的位置)、观察状态(与动作相同的位置信息)、来自三个摄像头的图像观察(每个摄像头分辨率为480x640,3通道彩色视频,使用av1编解码器),以及时间戳、帧索引、集索引和任务索引等元数据。数据以parquet文件格式存储,视频文件以mp4格式存储,总数据文件大小为100MB,视频文件大小为200MB。数据集适用于机器人学习和控制任务,但具体应用场景和目标任务未在README中明确说明。
This dataset is a robotics control dataset created using the LeRobot tool. It is designed for the so_follower robot type, containing 2 training episodes with a total of 933 frames, recorded at 30 fps. The dataset structure includes action data (e.g., positions of shoulder, elbow, wrist, and gripper), observation state (same position information as actions), image observations from three cameras (each with 480x640 resolution, 3-channel color video, using av1 codec), and metadata such as timestamp, frame index, episode index, and task index. Data is stored in parquet file format, with video files in mp4 format; total data file size is 100MB and video file size is 200MB. The dataset is suitable for robotics learning and control tasks, but specific application scenarios and target tasks are not explicitly detailed in the README.




