Four interconnected digital environments developed during the project — from public information to AI-powered learning tools.
The main public-facing platform for Ai-LEARN — project information, partners, deliverables, and links to all tools and resources. Updated continuously throughout the project.
Interactive browser-based tools for environmental modelling — flood forecasting using LSTM networks, groundwater data infilling with ANN models, and drought assessment using geostatistical methods. Powered by real EU satellite and sensor data.
An internationally accessible virtual learning environment built on IHE Delft's eCampus Moodle infrastructure. Supports MSc-level education, short courses, and professional development in English, Spanish, and Greek. WCAG 2.1 AA accessible.
An agentic AI teaching framework that personalises learning for students, supports teachers with pedagogical intelligence, and improves course quality over time.
AURA-Learn — Agentic Understanding, Reflection and Adaptation for Learning — is the project's flagship teaching innovation. Rather than a passive content delivery system, AURA-Learn creates an intelligent learning environment with three coordinated AI agents.
Each student is accompanied by a personal AI guide that provides exercises, examples, links, games, videos, and explanations adapted to their learning style and progress. At the same time, a teacher agent synthesises progress across the entire group, identifying common difficulties and informing the lecturer's pedagogical decisions. An independent evaluator agent monitors the course as a whole, comparing knowledge before and after, and feeding insights back into future course editions.
The result is a system where students receive personalised guidance, teachers receive pedagogical intelligence, and courses continuously improve through accumulated learning evidence.
Explore AURA-Learn in full →