Accurate mobility forecasting has wide-ranging implications for transportation systems, environmental management, and maritime safety. However, existing models often fall short in dynamic environments due to their limited integration of real-world constraints.
Gabriel Spadon from Dalhousie University, Canada, and Jose F. Rodrigues Jr. from the University of São Paulo (ICMC/USP), Brazil, are joining on a research project to address these gaps. Titled “Towards Optimized Mobility Forecasting Models with (Geo)Physics-Informed Machine Learning”, the initiative will develop advanced forecasting models informed by Canadian maritime data.
The project, funded by the National Council for Scientific and Technological Development (CNPq), will begin in early 2025. It will explore novel approaches, such as physics-informed neural networks (PINNs) and graph-based analysis, to better account for the interactions between moving entities and environmental factors.
Public information about this project’s acceptance is available at the funding agency website under the name “CNPq/MCTI/FNDCT Nº 22/2024 - Programa Conhecimento Brasil - Apoio a Projetos em Rede com Pesquisadores Brasileiros no Exterior”, online at https://bit.ly/4go5fnR.
To support this effort, a PhD position will be advertised at FCS/DAL and another at ICMC/USP. The positions will focus on model design, experimentation, benchmarking, fine-tuning, and scaling. During the project duration, the candidates will be able to conduct study exchanges among the participating institutions.
This collaboration brings together expertise from both institutions to address critical challenges in mobility modeling. Stay tuned for updates on this exciting initiative, including details about the PhD opportunity.
