Research Pillars

The Research Pillars of ENFIELD focus on advancing four key areas of AI—Green AI, Adaptive AI, Human-Centric AI, and Trustworthy AI—to address real-world challenges in Energy, Healthcare, Manufacturing, and Space. These pillars guide cutting-edge, interdisciplinary research through close collaboration between academia, research centers, and industry, ensuring impactful scientific results and knowledge sharing.

Mapping the Research Pillars

Research

Pillars

The Research Pillars was founded by 4 core dimensions:

Green AI
Adaptative AI
Human-Centric AI and
Trustworthy AI

Areas

The Research Pillars are structured into Research Areas:
13 active areas driving focused investigation and innovation.

Topics

The Research Areas are populated by Research Topics:
54 active topics driving specialized research and development.

Projects

TES Projects

Some Research Topics may be supported by Topics Exchange Schemes (TES) projects—not shown in the figure.
22 TES projects awarded

TES Projects Open Calls

Through TES Open Call Challenges scheme funded by cascading open calls in the context of the 4 pillars:
76 researchers engaged

Research Pillars

Green AI

Green AI approaches aim to minimize the carbon footprint while maintaining the performance of machine learning. The research under this pillar seeks to incorporate expert knowledge into machine learning models to improve data efficiency and enable sustainable training and operation of AI models.

Leader: SINTEF
Co-leader: KNOW
Research entities: IMT, TU/E, UCM, POLIMI, SINTEF, KNOW 
Industry players: PREDICT, TELENOR .

Adaptive AI 

Under the Adaptive AI pillar, research focuses on enhancing adaptability, efficiency, and reliability of AI systems in dynamic real-world environments by developing and accessing methods for adaptive AI on the edge and ensuring robustness and trustworthiness and drawing inspiration from the brain to study AI system adaptation. 
 
Leader: IMT
Co-leader: TU/E
Research entities: IMT, TU/E, UPB, DTI, NRS, UON, TUC, SINTEF, BME 
Industry players: LUSIADAS 

Human-Centric AI 

The Human-Centric AI pillar focuses on developing and evaluating methods and algorithms to enhance human-AI interaction in decision support, user experience, explainability, and interpretability, while also considering societal and ethical impacts. It aims to enhance transparency and human oversight in AI-based decision-making.

Leader: TU/E
Co-leader: UPB
Research entities: INESC, IMT, UPB, BME, DTI, NRS, TU/e, UON, TUC, LUSI, POLIMI, VTT
Industry players: TELENOR 

Trustworthy AI

The Trustworthy AI dimension focuses on developing models that promote responsible AI, enhancing user trust, adoption, and acceptance. Formal design approaches for abstraction and uncertainty modelling in AI will be revisited to create reliable tools. Research will also advance in specifications, robust learning methods, and improving data quality in AI systems.

Leader: NTNU
Co-leader: TUC
Research entities: NTNU, TUC, BME, DTI, UON, VTT, INESC TEC, ICCS 
Industry players: MAGGIOLI, TELENOR 

Contribution to environmental, social and governance objectives

Economic and Technological Impact
The research within the pillars will train future AI leaders and enhance industry competitiveness by applying AI to real-world use cases, contributing to Sustainable Development Goals 4, 8, 9, and 17.

Health and Safety
Advance Trustworthy, Human-Centric AI to increase adoption in healthcare, ensuring data privacy and reducing biases, supporting Sustainable Development Goal 3.

Environment
Foster EU research collaboration between academia and industry through practical use cases in Green AI, ensuring applicable results across industries and supporting Sustainable Development Goals 7, 9, 11, 12, and 13.