AI research

Detecting irregularities – The CausalAnomalies project

Anomaly detection is essential for several areas of application at the German Aerospace Center (Deutsches Zentrum für Luft- und Raumfahrt; DLR). Anomaly detection plays a crucial role, for example, when monitoring satellite telemetry or analysing Earth observation data from satellites and climate models. Similarly, the continuous and automated monitoring of the condition of rails and railway switches is important to enable timely maintenance of transport infrastructure. In aviation, data is also continuously collected and used to monitor aircraft systems.

Another potential area of application is the examination of log data in information and communications technology (ICT) systems. Here, suspicious usage and attack patterns can be detected in locations such as the log files produced during the flight of an uncrewed aircraft. In the event of deviations from expected patterns, users can intervene to stop or adjust the procedures. When an anomaly is detected, it can indicate that the monitored system has entered a certain state, such as in the event of damage, which can be recognised and remedied through such intervention. However, understanding the causes of such anomalies is particularly important to model and predict the necessary procedures before the state of the system changes.

The CausalAnomalies project aims to develop new, causally explicable algorithms that detect anomalies. Finding abnormal states in time series data is a goal whenever sequential data is recorded and analysed. These data are often high-dimensional, error-prone, incomplete or not labelled. Furthermore, it is frequently unclear which abnormal behaviour within the data is relevant. Therefore, the first project goal is to explore different AI methods for anomaly detection. The next goal is to make these methods explicable in order to overcome the 'black box' nature of many AI methods. For this purpose, causal methods will be used and adapted. The developed methods will then be applied to various use cases at DLR and will be tested and optimised.

These use cases include:

  • Anomaly detection in satellite telemetry data
  • Detection of regimes of causal relationships in Earth system model data
  • Detection of anomalies and attack patterns in ICT systems
  • Condition monitoring of control and safety systems in transport infrastructure
  • Identification of causes of abnormal behaviour or incorrect decision making in the log files of uncrewed aircraft
  • Extension of condition monitoring functions for aircraft systems

The CausalAnomalies project is part of DLR’s research in 'Artificial intelligence'.


Tobias Schneiderhan

Acting Board Member for Digitalization
German Aerospace Center (DLR)
Linder Höhe, 51147 Cologne