AFKUI

Analysesysteme zur Früherkennung von Bedrohungen an Kritischen Unterseeischen Infrastrukturen

underwater cables and a submarine

The AFKUI Project

Analysis system for early detection of threats to critical undersea infrastructures

Cables and pipelines under water are being damaged more and more often – on purpose. Yet, this kind of infrastructure is of the utmost importance for the EU. Two EU directives now require member states to identify critical infrastructure and ensure they are more resilient to attacks, sabotage or natural hazards.

Office

Project Specs

  • Timeframe:  May 1, 2025 – April 30, 2028
  • Funding: €3.3 million
  • Goal:  Improve the protection and monitoring of critical undersea infrastructure.
afkui-small

Project Summary

AFKUI is a project funded by the BMWE and aims to develop new intelligent analysis systems for early detection and management of threat situations to critical undersea infrastructure. To achieve this, open source intelligence data sources will be collected and combined with commercially available information. The data then correlates and detects threats early on with the help of modern AI algorithms.

BMWE-en

Funding

Funding is provided by the Federal Ministry for Economic Affairs and Energy.

 

Our Contribution

Within the AFKUI project, north.io is responsible for accelerated computing and developing and providing a data warehouse with the according API for data collection and rendering. This is necessary to analyse the data and detect potential threats. First, we evaluated relevant algorithms to be tested for efficiency gains on the infrastructure for accelerated computing. After several tests, these algorithms are implemented and further tests with sensor data are executed. The Data Warehouse is then implemented and test data for all relevant formats uploaded. We then provide the project partners with the access to the Data Warehouse and supply interfaces for input and output data.

critical infrastructure underwater

Consortium

North Logo IBM_logo.svg DLR-Logo