Research Areas
To achieve the proposed objective, the research posed in AI4Transport focuses on the following areas, because of their impact on the aforementioned challenges of urban mobility and freight transport:
- New traffic prediction methods that improve current systems by reducing the need for expert knowledge for the generation of prediction models and by extending their spatial and temporal scope. To this end, the aim is to develop new Automatic Machine Learning and Deep Learning techniques.
- New tools for Travel Behaviour Analysis that contribute to a better understanding of how citizens use and will use emerging modes of mobility. For this purpose, the design of Machine Learning and Data Fusion techniques capable of handling multiple heterogeneous data sources simultaneously is proposed.
- New models and optimization algorithms that contribute to a more collaborative and synchromodal logistics. More specifically, the aim is first to create richer versions of the Vehicle Routing Problem that consider collaboration between agents and that model the use and particularities of emerging mobility modes, and optimization methods based on hybrid metaheuristics capable of addressing the complexity of these models.
