A Smartphone GPS and Digital Map-Based Deep Learning Method for Traffic Stream Forecasts
スマホプローブおよびデジタル地図を活用した深層学習による自動車交通流予測
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.)：￥1,100 Members (tax incl.)：￥880
- Publication code
- 20244464
- Paper/Info type
- Journal of Society of Automotive Engineers of Japan
Vol.78 No.7
- Pages
- 118-124(Total 7 p)
- Date of publication
- Jul 2024
- Publisher
- JSAE
- Language
- Japanese
Detailed Information
| Category(J) | ホットトピックス Translation |
|---|---|
| Category(E) | Hot Topics |
| Author(J) | 1) 池田 真土里 |
| Author(E) | 1) Madori Ikeda |
| Affiliation(J) | 1) ジオテクノロジーズ |
| Abstract(J) | 自動車交通流の予測は，その重要性にも関わらず交通流捕捉の難しさや道路構造の複雑さのため，幹線道路に限定され時間解像度や即応性が低いものが多い．我々は，スマホプローブから交通流を捕捉し，複雑な構造を扱える深層学習技術を用いて，生活道路も含む多くの道路上の交通流について数分単位の予測を即応的に行う． Translation |
| Abstract(E) | Despite their importance, the present traffic stream forecasts are applied only to arterial roads without immediate responsiveness, and their time resolution is also often low. There are due to the difficulty of capturing traffic streams and the complexity of road structures. We propose a quick response method which provides more types of roads with high time resolution forecasts. The method realizes such a forecast by capturing real-time traffic streams on the roads from smartphones with GPS tracking devices and by using recent deep learning techniques which can performs time series forecasts with handling the complex structures successfully. |
