Improved real-time velocity prediction by considering preceding vehicle dynamics

Haidi Sun, Junqiu Li, Chao Sun

科研成果: 书/报告/会议事项章节会议稿件同行评审

1 引用 (Scopus)

摘要

This paper focuses on improving the previous velocity prediction method performance by incorporating preceding vehicle dynamics. Firstly, a vehicle-following system is established to obtain the target vehicle velocity, preceding vehicle velocity and the distance between them. After a systematic correlation analysis, an Artificial Neural Network (ANN) based on real-time velocity prediction is proposed to improve the prediction accuracy regarding the previous studies in the literature. The interaction pattern between front vehicle and target vehicle is learnt via the ANN model. Simulation results indicate that the improvement mainly gains from the awareness of acceleration switching dynamics during driving. The proposed method is able to increase prediction accuracy by over 30%. The velocity predictor can be used in the energy management, safety control or other fields for automotive engineering.

源语言英语
主期刊名2019 IEEE Vehicle Power and Propulsion Conference, VPPC 2019 - Proceedings
出版商Institute of Electrical and Electronics Engineers Inc.
ISBN(电子版)9781728112497
DOI
出版状态已出版 - 10月 2019
活动2019 IEEE Vehicle Power and Propulsion Conference, VPPC 2019 - Hanoi, 越南
期限: 14 10月 201917 10月 2019

出版系列

姓名2019 IEEE Vehicle Power and Propulsion Conference, VPPC 2019 - Proceedings

会议

会议2019 IEEE Vehicle Power and Propulsion Conference, VPPC 2019
国家/地区越南
Hanoi
时期14/10/1917/10/19

指纹

探究 'Improved real-time velocity prediction by considering preceding vehicle dynamics' 的科研主题。它们共同构成独一无二的指纹。

引用此