Highway travel speed prediction based on ETC toll data

Wei Liu; Min Zhao

doi:10.1117/12.2645856

23 November 2022 Highway travel speed prediction based on ETC toll data

Wei Liu, Min Zhao

Proceedings Volume 12302, Seventh International Conference on Electromechanical Control Technology and Transportation (ICECTT 2022); 1230230 (2022) https://doi.org/10.1117/12.2645856
Event: Seventh International Conference on Electromechanical Control Technology and Transportation (ICECTT 2022), 2022, Guangzhou, China

Abstract

The accuracy of traffic flow prediction is significantly degraded by data anomalies and data noise. To solve this problem, a hybrid traffic flow prediction model based on iForest-VMD-GRU is proposed in this paper. Firstly, we detect the outlier anomalies in the original traffic speed sequence through iForest and use interpolation to complete the normal traffic speed sequence. Then, to reduce the interference of noisy data and improve the model prediction accuracy, we decompose the normal traffic speed sequence into basic trend components and multiple random fluctuation components through VMD. Finally, GRU is used to predict each subsequence, and the predicted values of each subsequence are combined into the final prediction results. The empirical analysis shows that the proposed model in this paper can significantly improve the prediction performance.

Citation Download Citation

Wei Liu and Min Zhao "Highway travel speed prediction based on ETC toll data", Proc. SPIE 12302, Seventh International Conference on Electromechanical Control Technology and Transportation (ICECTT 2022), 1230230 (23 November 2022); https://doi.org/10.1117/12.2645856

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available