Authors: Matthieu Dolbeault, Olga Mula, Agustín Somacal
Published on: February 05, 2024
Impact Score: 8.2
Arxiv code: Arxiv:2402.02812
Summary
- What is new: The introduction of hybrid methods that combine data-driven and physics-driven approaches for urban air pollution map reconstruction.
- Why this is important: Reconstructing real-time urban air pollution maps is challenging due to data heterogeneity, scarcity of measurements, noise, and large areas involved.
- What the research proposes: Using city graphs and employing fully data-driven, physics-driven, and hybrid strategies combined with super-learning models.
- Results: The methods showed promising results when tested in the inner city of Paris, showing that reliable air pollution maps can be generated in real-time.
Technical Details
Technological frameworks used: Super-learning models, city graphs
Models used: Data-driven models, physics-driven models, hybrid models
Data used: Direct and indirect air pollution measurements
Potential Impact
Environmental monitoring firms, urban planning agencies, companies developing smart city solutions, public health organizations
Want to implement this idea in a business?
We have generated a startup concept here: AirQualiMap.
Leave a Reply