AI4Realnet
AI For Real-World Network Operation Artificial IntelligenceBioengineeringPower and Energy Systems
Description
Artificial intelligence (AI) has the potential to enhance decision-making in the management of critical systems that have traditionally been under human direction. The EU-funded project, AI4REALNET will develop methods to prioritise trustworthiness in AI-assisted human control, incorporating augmented cognition, hybrid human-AI co-learning and autonomous AI, all while maintaining a focus on the resilience, safety and security of critical infrastructures. The project will also expedite the development and validation of new AI algorithms by the consortium and the broader AI community.
It will do this by leveraging open-source AI-friendly digital environments capable of simulating realistic scenarios involving the operation of physical systems and human decision-making. AI4REALNET will contribute to address the critical aspects of decarbonisation, digitalisation and resilience in power grids, railway and air traffic control.
Scientific Advances
The main contributions of AI4REALNET will be:
1) AI algorithms are mainly composed of reinforcement and supervised learning, unifying the benefits of existing heuristics, physical modeling of these complex systems and learning methods, and a set of complementary techniques to enhance transparency, safety, explainability, and human acceptance.
2) Human-in-the-loop decision-making that promotes co-learning between AI and humans, considering integration of model uncertainty, human cognitive load, and trust.
3) Autonomous AI systems rely on human supervision, integrating human domain knowledge and safety rules.
It will do this by leveraging open-source AI-friendly digital environments capable of simulating realistic scenarios involving the operation of physical systems and human decision-making. AI4REALNET will contribute to address the critical aspects of decarbonisation, digitalisation and resilience in power grids, railway and air traffic control.
Scientific Advances
The main contributions of AI4REALNET will be:
1) AI algorithms are mainly composed of reinforcement and supervised learning, unifying the benefits of existing heuristics, physical modeling of these complex systems and learning methods, and a set of complementary techniques to enhance transparency, safety, explainability, and human acceptance.
2) Human-in-the-loop decision-making that promotes co-learning between AI and humans, considering integration of model uncertainty, human cognitive load, and trust.
3) Autonomous AI systems rely on human supervision, integrating human domain knowledge and safety rules.
