Parameter calibration method for car-following models under snowy weather conditions: Integrating an informer time series encoder and physics-informed neural networks
Yaping Sun, Wenfang Li, Mengyang Yang, Xingchen Zhang · 2026
Read Parameter calibration method for car-following models under snowy weather conditions: Integrating an informer time series encoder and physics-informed neural networks by Yaping Sun, Wenfang Li, Mengyang Yang, Xingchen Zhang (2026) 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/plos-pone-0350550.txt.
Description
Yaping Sun and colleagues present a tuning method for car following models when roads turn snowy. Drivers space out differently on slick, low visibility roads, and standard models miss that shift. Five classic following models were tuned with road grip and visibility counted in. A time series encoder extracts snow scene features, while a physics guided network adjusts model settings on the fly.
Checked against a national driving data set and real snow driving tests, tuning accuracy beat a standard genetic search by about thirteen percent in snow. Gains held across all five model types. Better snow tuned models can sharpen traffic simulations, driver aids, and safety tests for winter roads.
Catalog record
| CREATOR | Yaping Sun, Wenfang Li, Mengyang Yang, Xingchen Zhang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55234 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |