Checkpoint data-driven GCN-GRU vehicle trajectory and traffic flow prediction

Deyong Guan, Na Ren, Ke Wang, Qi Wang, Hualong Zhang · 2024

Read Checkpoint data-driven GCN-GRU vehicle trajectory and traffic flow prediction by Deyong Guan, Na Ren, Ke Wang, Qi Wang, Hualong Zhang (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-11621339.txt.

Description

Deyong Guan and co-authors (2024) predict vehicle paths and traffic flow from checkpoint records instead of GPS traces. Checkpoints cover whole districts, are reliable, and weigh less than continuous tracking. The team turned Qingdao checkpoint data into daily vehicle journeys. A graph network learned the road layout's spatial patterns from how checkpoints connect. A gated recurrent unit then learned the time patterns in each journey. The path forecasts fed a flow forecast for every checkpoint.

The combined model beat each single model and two close variants on both error measures. Flow forecasts matched real counts with a percentage error of 0.18. The authors argue the order matters: predict where vehicles go first, then count them. City managers could use the flow forecasts for route planning support and early congestion warnings. The method leans on data cities already collect at toll and camera points, avoiding new sensors.

Catalog record

CREATORDeyong Guan, Na Ren, Ke Wang, Qi Wang, Hualong Zhang
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE50 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC11621339

Browse the full library · About the Tower of Babel