Spatio-temporal envolutional graph neural network for traffic flow prediction in UAV-based urban traffic monitoring system
Wenming Ma, Zihao Chu, Hao Chen, Mingqi Li · 2024
Read Spatio-temporal envolutional graph neural network for traffic flow prediction in UAV-based urban traffic monitoring system by Wenming Ma, Zihao Chu, Hao Chen, Mingqi Li (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11538314.txt.
Description
Wenming Ma, Zihao Chu, Hao Chen, and Mingqi Li propose a new neural network for predicting traffic flow in cities watched by drones. Their 2024 paper explains that drone-based monitoring feeds data rich in both space and time patterns, which graph neural networks handle well. Such networks, though, tend to smooth data too much as they deepen, which blurs the very structure they should capture.
An earlier model fought that smoothing with differential equations, but it fixed the map of which road points count as neighbors in advance. Traffic does not sit still, so the map should not either. The new model keeps redrawing that map during training, letting it follow the data's changing relations. Tests on several real-world data sets show the approach beating leading benchmarks. The authors argue that a map which evolves with the data is the key gain.
Catalog record
| CREATOR | Wenming Ma, Zihao Chu, Hao Chen, Mingqi Li |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 89 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |