Fujitsu wins first prize
10/03/2020
Airbus SE has awarded Fujitsu first prize in its Airbus AiGym competition. The global aerospace manufacturer’s competition was devised to pinpoint the most accurate unsupervised predictive artificial intelligence (AI) capability for helicopters.Flight engineers attach large numbers of sensors to test helicopters to capture every hint of behaviour. With vast amounts of data, early warning signals can be difficult to detect. The challenge was designed to support research into accurately locating issues – especially data outliers.
Fujitsu came out top out of 140 participants for developing a new way to use unsupervised AI to detect anomalies in accelerometer data from Airbus precertification helicopters. By leveraging the ‘DeepTAN’ Unsupervised AI Model created by its subdivision, Fujitsu Systems Europe (FSE), the solution achieved 93% precision.
The lauded design took data sequences from multiple sensors and analysed them across a fixed time period, detecting abnormal sensor behaviour using a deep learning algorithm based on multivariate anomaly detection with generative adversarial networks (MAD-GEN). FSE both trained and validated the algorithm using 1677 one-minute sequences of accelerometer data from test helicopters flying at various locations and angles.
