Home Bots in Society AI improves sustainability and security of critical infrastructure

AI improves sustainability and security of critical infrastructure

by Pieter Werner

Artificial Intelligence can support human decisions, increase efficiency and security in the operation of critical infrastructure. This is the aim of the European Horizon Europe project AI4REALNET – AI for REAL-World network operation. Politecnico di Milano is among the partners of this project that supports operators of electrical, railway and air traffic systems in the implementation of artificial intelligence-human interactions.

Led by the Portuguese research institute INESC TEC and involving the Department of Electronics, Information and Bioengineering and the Department of Management, Economics and Industrial Engineering of Politecnico di Milano and other partners from France, Germany, the Netherlands, Switzerland, Sweden and Austria, the project promotes collaboration between humans and artificial intelligence. The goal is to support the decisions of human operators and to create the conditions for the decarbonisation of crucial sectors such as energy and transport.

It is not about replacing humans with AI, but rather ensuring that AI emerges as a support for faster decisions and even planning specific tasks autonomously. The project aims to reduce the operators’ workload in areas where human intervention prevails, offering opportunities to address critical infrastructure challenges. AI4REALNET developments will be validated in six industrial use cases, demonstrating tangible added value.

‘The ultimate goal ,’ explains Prof. Marcello Restelli, project coordinator for Politecnico di Milano, ‘is to improve the security and resilience of critical infrastructure, which is becoming increasingly complex due to the increase in information and the challenges imposed by decarbonisation. The AI4REALNET consortium believes that AI can increase operational effectiveness and reduce errors’.

Industry involvement will promote awareness of the benefits of reinforcement learning and explainable machine learning. The project will also use open-source AI-friendly digital environments such as Grid2Op, Flatland and BlueSky to foster and advance a global artificial intelligence community.

The project, which is funded with around 4 million euros by the European Union through the Horizon Europe programme and 2 million euros by the Swiss State Secretariat for Education, Research and Innovation (SERI), emphasises international collaboration to tackle crucial societal challenges.

Misschien vind je deze berichten ook interessant