Forgis/FactoryNet_Dataset
收藏资源简介:
FactoryNet是一个大规模、多机器的工业数据集,专为工业机器人和制造异常检测设计。该数据集通过将来自完全不同机器的高频物理数据统一到一个数学上统一的坐标系中,解决了工业数据集分散的问题。数据集包含来自三种不同开源工业数据集的数据:UMich CNC Mill Tool Wear Dataset(3轴CNC铣床)、AURSAD(UR3e 6轴协作机器人)和voraus-AD(Yu-Cobot 6轴协作机器人)。每种机器都有不同的任务和异常类型。数据集提供了原始值和标准化值两种配置,便于不同用途的研究。标准化模式通过统一前缀命名和轴索引,使得跨机器学习成为可能。
FactoryNet is a large-scale, multi-machine industrial dataset designed for industrial robotics and manufacturing anomaly detection. It solves the problem of siloed industrial datasets by unifying high-frequency physical data from completely different machines into a single, mathematically unified coordinate system. The dataset includes data from three distinct open-source industrial datasets: UMich CNC Mill Tool Wear Dataset (3-Axis CNC Mill), AURSAD (UR3e 6-Axis Collaborative Robot), and voraus-AD (Yu-Cobot 6-Axis Collaborative Robot). Each machine has different tasks and anomaly types. The dataset is available in two configurations: raw values and normalized values, facilitating research for different purposes. The standardized schema, with unified prefix naming and axis indexing, enables cross-machine learning.




