Truck speed prediction with a simple two-stage neural network using GPS data and Monte Carlo (MC) dropout

Yuze Zhang; Jiangtao Li; Lei Deng; Yu Zang; Yucen Zhang

doi:10.1117/12.2689655

7 September 2023 Truck speed prediction with a simple two-stage neural network using GPS data and Monte Carlo (MC) dropout

Yuze Zhang, Jiangtao Li, Lei Deng, Yu Zang, Yucen Zhang

Author Affiliations +

Proceedings Volume 12790, Eighth International Conference on Electromechanical Control Technology and Transportation (ICECTT 2023); 1279057 (2023) https://doi.org/10.1117/12.2689655
Event: 8th International Conference on Electromechanical Control Technology and Transportation (ICECTT 2023), 2023, Hangzhou, China

Abstract

According to statistics, large heavy-haul truck is the main vehicle type causing serious traffic accidents in China and speeding is one of the most common cases. It is necessary to study the travel speed prediction of trucks and provide early warnings for drivers as well as manage departments in previous to serious traffic accidents. Based on the GPS generated by trucks driving in China, a simple two-stage neural network was proposed in this paper to predict the travel speed of trucks with Monte Carlo Dropout. The proposed network was composed of an LSTM Network and a Fully Connected Feedforward Neural Network. The MC Dropout was utilized as an optimum approach for improving the predictions. Accuracies of the proposed network were tested with real trajectories and the RMSE was about 1.61 km/h with a bias of around -0.57 km/h.

Citation Download Citation

Yuze Zhang, Jiangtao Li, Lei Deng, Yu Zang, and Yucen Zhang "Truck speed prediction with a simple two-stage neural network using GPS data and Monte Carlo (MC) dropout", Proc. SPIE 12790, Eighth International Conference on Electromechanical Control Technology and Transportation (ICECTT 2023), 1279057 (7 September 2023); https://doi.org/10.1117/12.2689655

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