Oihane Gómez-Carmona and Erik Eguskiza-Aranda participated in the 17th International Conference on Ubiquitous Computing and Ambient ‪Intelligence UCAmI 2024, held in Florence, Italy, in November 2025. The UCAmI Conference focuses on Ambient Intelligence (AmI), a field that promotes user-centred computing environments and systems aimed at seamlessly integrating information technologies into everyday devices and activities.

There, Oihane and Erik presented the achievements and ongoing work of the NextEdge project, developed in collaboration with Next Mobility Solutions and its CEO Dr. Javier Goikoetxea Gonzalez. This initiative aims to build an intelligent co-pilot for connected mobility by combining hybrid Edge-Cloud processing with advanced AI techniques. Their contributions showcased progress toward systems capable of anticipating mobility needs, supporting safer and more efficient mobility decisions, and generating adaptive, AI-driven user insights.

On the one hand, Oihane Gómez presented the work “Towards Adaptive and Personalized Co-Pilot Systems: Design Principles for Context-Aware Driver Assistance”, which introduces a set of foundational design principles for next-generation driver-assistance copilots. Her contribution focuses on how adaptive and personalized systems can leverage contextual information to enhance user experience, improve safety, and support real-time decision-making. The work emphasizes the importance of tailoring assistance to individual drivers and situational factors, paving the way for more intelligent, transparent, and human-centered in-vehicle support technologies.

On the other hand, Erik Eguskiza presented the work “From Rules to Representations: Improving Event Classification with Transformers to Infer Mobility Needs”, which investigates how machine learning and transformer-based language models can infer whether calendar events require physical transportation from their textual descriptions. His contribution examines a hybrid dataset generation strategy combining real and synthetic events, evaluates multiple classification approaches, including traditional ML models and DistilBERT, and demonstrates the advantages of contextual language models in identifying implicit mobility needs. The study provides a scalable methodology for mobility-aware intelligent assistants, enabling applications such as automated route planning and smart agenda management.

This work has been funded by the Ministry of Industry and Tourism through the PERTE VEC II as PROYECTO ITAM for NEXT MOBILITY SOLUTIONS under the “expediente VE2-010000-2023-75.