R&I group

Distributed AI

The Distributed Artificial Intelligence (DAI) research group conducts research and innovation activities in explainable and distributed AI, multimodal AI perception, and cognitive cloud. The group has a long track record of participation in EU R&D projects, Catalan and Spanish-funded research and innovation projects, and bilateral collaborations with private industry and the public sector. The research group develops and continuously updates a research roadmap to steer its activities and identify new fields to explore. One of the key added values of the DAI research group is its mission to boost the overall capabilities of i2CAT in AI, fostering synergies with the centre’s other research groups.

Research lines

Trustworthy AI
- Human Centric AI: Explainable and GenAI
- Decentralized Collaborative AI
- Quantum AI

The “Trustworthy AI” research line focuses on developing AI systems that are reliable, ethical, and sustainable. This includes Human-Centric AI, emphasizing Explainable AI and responsible Generative AI . We also explore Decentralized AI for secure and scalable distributed systems. Furthermore, we delve into Quantum AI, investigating how quantum computing can enable high-dimensional and efficient AI algorithms.

Multimodal AI-Perception
- Computational Vision and Scene Analysis
- NLP – Acoustic AI
- Volumetric Synthesis
Distributed Data Architectures
- AI-based DDA optimizations
- Data efficiency and privacy optimizations

In this research line we focus on complementing our AI technologies with cutting edge research on Distributed Data Architectures and optimizations. We specifically work on novel AI-centric data efficiency and privacy optimizations, such as data anonymization, aggregation and compression. We also delve into AI-based data optimizations, particularly for distributed data lake and edge-cloud management systems.

Innovation lines

Artificial Intelligence
Data Spaces

Technologies

Predictive models for risk detection from data logs
Enhanced explainability models for production environments
Synthetic data generation with GenAI
Deep Reinforcement Learning for Swarm Systems
Multi-agents systems for optimization problems
Audio classification with Deep Learning methods from cloud to Tiny ML
Multi-detection and multi-tracking of objects with multi-modal Computer Vision
Deep Learning for calibration, compression, recognition and generation of volumetric content
Data Spaces Components based on IDSA and Gaia-X architectures
Cognitive cloud optimizations based on AI
Stretched distributed and federated Data Lakes
Data efficiency and privacy optimizations
RAG for multimodal Large AI Models

Group leader

Miguel Angel Veganzones

Areas

Smart Networks & Services 5G/6G Space Communications Immersive & Interactive Multimedia Technologies Artificial Intelligence (AI)

Distributed AI

Projects

Distributed AI

Publications

Interactive digital twins enabling responsible extended reality applications. (Nature, Scientific Reports)

Antoun, M., Elhajj, I.H., Sayour, M. et al. Interactive digital twins enabling responsible extended reality applications. Sci Rep 15, 34539 (2025). https://doi.org/10.1038/s41598-025-17855-9

Considerations, challenges and opportunities when developing data-driven models for process manufacturing systems

Oliver J Fisher, Nicholas J Watson, Josep E Escrig, Rob Witt, Laura Porcu, Darren Bacon, Martin Rigley, Rachel L Gomes, “Considerations, challenges and opportunities when developing data-driven models for process manufacturing systems”, Computers & Chemical Engineering, Volume 140, 2020. https://www.sciencedirect.com/science/article/abs/pii/S0098135419308373

Ultrasonic Measurements and Machine Learning for Monitoring the Removal of Surface Fouling during Clean-in-Place Processes

J. Escrig, E. Woolley, A. Simeone, N.J. Watson, “Monitoring the cleaning of food fouling in pipes using ultrasonic measurements and machine learning”, Food Control, Volume 116, 2020,

Monitoring the cleaning of food fouling in pipes using ultrasonic measurements and machine learning

Escrig, J., Woolley, E., Simeone, A., & Watson, N. J. (2020). Monitoring the cleaning of food fouling in pipes using ultrasonic measurements and machine learning. Food Control, 107309. https://doi.org/10.1016/j.foodcont.2020.107309

RBD24 : A Labelled Dataset with Risk Activities Using Log Application Data

Social media as a source of drug safety information in the pediatric population

view all

Distributed AI