voraus-AD
收藏资源简介:
该数据集名为“voraus-AD”,包含了协作机器人执行工业取放任务时的机器数据。其中既包含了用于训练的正常数据,也包含了用于测试的正常数据和异常数据,涵盖了12种不同的异常类型。该数据集共记录了130个信号,采样频率为500赫兹,全面覆盖了机械臂操作的各种方面。它旨在用于训练异常检测模型,并评估如MVT-Flow等方法。在取放操作过程中,共记录了2122个时间序列样本,该数据集的任务是针对机器人应用中的异常检测。
This dataset, named "voraus-AD", contains machine data collected when collaborative robots perform industrial pick-and-place tasks. It includes normal data for model training, as well as both normal and abnormal data for model testing, covering 12 distinct anomaly types. A total of 130 signals were recorded at a sampling frequency of 500 Hz, comprehensively covering all aspects of robotic arm operations. This dataset is intended for training anomaly detection models and evaluating methods such as MVT-Flow. During the pick-and-place operations, 2122 time-series samples were recorded in total, and the core task of this dataset is anomaly detection in robotic applications.

- 1The voraus-AD Dataset for Anomaly Detection in Robot Applications汉诺威莱布尼兹大学 · 2023年



