Projektbeschreibung
KI-basierte Entscheidungsfindung hilft beim Betrieb kritischer Systeme
Künstliche Intelligenz (KI) kann ein leistungsfähiges Instrument für den Betrieb kritischer Systeme sein, die ursprünglich von Menschen kontrolliert wurden. Im Rahmen des EU-finanzierten Projekts AI4REALNET werden Methoden entwickelt, um das Vertrauen in KI-gestützte menschlicher Kontrolle zu priorisieren und dabei erweiterte Kognition, hybrides Lernen zwischen Mensch und KI sowie autonome KI einzubeziehen, wobei der Schwerpunkt auf der Resilienz und Sicherheit kritischer Infrastrukturen liegt. Das Projektteam wird außerdem die Entwicklung und Validierung neuer KI-Algorithmen durch das Konsortium und die breitere KI-Gemeinschaft beschleunigen. Dazu werden KI-freundliche frei zugängliche digitale Umgebungen genutzt, die realistische Szenarien simulieren können, die den Betrieb physischer Systeme und menschliche Entscheidungen einschließen. Schließlich wird AI4REALNET dazu beitragen, die wesentlichen Aspekte der Dekarbonisierung, Digitalisierung und Widerstandsfähigkeit anzugehen.
Ziel
The scope of AI4REALNET covers the perspective of AI-based solutions addressing critical systems (electricity, railway, and air traffic management) modelled by networks that can be simulated, and are traditionally operated by humans, and where AI systems complement and augment human abilities. It has two main strategic goals: 1) to develop the next generation of decision-making methods powered by supervised and reinforcement learning, which aim at trustworthiness in AI-assisted human control with augmented cognition, hybrid human-AI co-learning and autonomous AI, with the resilience, safety, and security of critical infrastructures as core requirements, and 2) to boost the development and validation of novel AI algorithms, by the consortium and AI community, through existing open-source digital environments capable of emulating realistic scenarios of physical systems operation and human decision-making.
The core elements are: a) AI algorithms mainly composed by supervised and reinforcement learning, unifying the benefits of existing heuristics, physical modelling of these complex systems and learning methods, as well as, a set of complementary techniques to enhance transparency, safety, explainability and human acceptance; b) human-in-the-loop decision making for co-learning between AI and humans, considering integration of model uncertainty, human cognitive load and trust; c) autonomous AI systems relying on human supervision, embedded with human domain knowledge and safety rules.
The AI4REALNET framework will be validated in 6 uses cases driven by industry requirements, across 3 network infrastructures with common properties. The use cases are focused on critical challenges and tasks of network operators, considering strategic long-term goals, such as decarbonisation, digitalisation, and resilience to disturbances, and are formulated in a unified sequential decision problem where many AI and non-AI algorithms can be applied and benchmarked.
Wissenschaftliches Gebiet
- natural sciencescomputer and information sciencesartificial intelligencemachine learningreinforcement learning
- engineering and technologycivil engineeringtransportation engineering
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectrical engineeringpower engineering
- social sciencessocial geographytransporttransport planningair traffic management
- natural sciencescomputer and information sciencesartificial intelligenceheuristic programming
Schlüsselbegriffe
Programm/Programme
Thema/Themen
Aufforderung zur Vorschlagseinreichung
Andere Projekte für diesen Aufruf anzeigenFinanzierungsplan
HORIZON-RIA - HORIZON Research and Innovation ActionsKoordinator
4200 465 Porto
Portugal