Trajectory prediction method using deep learning for intelligent and connected vehicles

Tianqi Qie; Weida Wang; Chao Yang; Ying Li; Yuhang Zhang; Wenjie Liu

doi:10.1109/ICPS58381.2023.10128049

Trajectory prediction method using deep learning for intelligent and connected vehicles

Tianqi Qie, Weida Wang, Chao Yang, Ying Li, Yuhang Zhang, Wenjie Liu

School of Mechanical Engineering

Beijing Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

The trajectory prediction is significant for the driving safety of intelligent and connected vehicles. To accurately predict the vehicle trajectory, a hybrid method combining physic-based and data-based methods is proposed for intelligent and connected vehicles. The proposed method applied the physic-based method to represent vehicle kinematics. Then, the error of the physic-based method, which is the unmodeled features, is modeled with the data-based deep learning method using Encoder-Decoder Long short-term memory (LSTM). The proposed method is trained and evaluated by an actual vehicle dataset. When the prediction horizon is 3s, compared with the physic-based method, the longitudinal error, lateral error, and yaw angle error decreased by 93.9%, 86.6%, and 76.0%, respectively. Results show that the proposed method improves the trajectory prediction accuracy of autonomous and connected vehicles.

Original language	English
Title of host publication	Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798350311259
DOIs	https://doi.org/10.1109/ICPS58381.2023.10128049
Publication status	Published - 2023
Event	6th IEEE International Conference on Industrial Cyber-Physical Systems, ICPS 2023 - Wuhan, China Duration: 8 May 2023 → 11 May 2023

Publication series

Name	Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023

Conference

Conference	6th IEEE International Conference on Industrial Cyber-Physical Systems, ICPS 2023
Country/Territory	China
City	Wuhan
Period	8/05/23 → 11/05/23

Keywords

intelligent and connected vehicles
long short-term memory (LSTM)
trajectory prediction

Access to Document

10.1109/ICPS58381.2023.10128049

Cite this

Qie, T., Wang, W., Yang, C., Li, Y., Zhang, Y., & Liu, W. (2023). Trajectory prediction method using deep learning for intelligent and connected vehicles. In Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023 (Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICPS58381.2023.10128049

Qie, Tianqi ; Wang, Weida ; Yang, Chao et al. / Trajectory prediction method using deep learning for intelligent and connected vehicles. Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023. Institute of Electrical and Electronics Engineers Inc., 2023. (Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023).

@inproceedings{9c0cc3c15fc44c99a480000ae64e3113,

title = "Trajectory prediction method using deep learning for intelligent and connected vehicles",

abstract = "The trajectory prediction is significant for the driving safety of intelligent and connected vehicles. To accurately predict the vehicle trajectory, a hybrid method combining physic-based and data-based methods is proposed for intelligent and connected vehicles. The proposed method applied the physic-based method to represent vehicle kinematics. Then, the error of the physic-based method, which is the unmodeled features, is modeled with the data-based deep learning method using Encoder-Decoder Long short-term memory (LSTM). The proposed method is trained and evaluated by an actual vehicle dataset. When the prediction horizon is 3s, compared with the physic-based method, the longitudinal error, lateral error, and yaw angle error decreased by 93.9%, 86.6%, and 76.0%, respectively. Results show that the proposed method improves the trajectory prediction accuracy of autonomous and connected vehicles.",

keywords = "intelligent and connected vehicles, long short-term memory (LSTM), trajectory prediction",

author = "Tianqi Qie and Weida Wang and Chao Yang and Ying Li and Yuhang Zhang and Wenjie Liu",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 6th IEEE International Conference on Industrial Cyber-Physical Systems, ICPS 2023 ; Conference date: 08-05-2023 Through 11-05-2023",

year = "2023",

doi = "10.1109/ICPS58381.2023.10128049",

language = "English",

series = "Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023",

address = "United States",

}

Qie, T, Wang, W , Yang, C , Li, Y, Zhang, Y & Liu, W 2023, Trajectory prediction method using deep learning for intelligent and connected vehicles. in Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023. Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023, Institute of Electrical and Electronics Engineers Inc., 6th IEEE International Conference on Industrial Cyber-Physical Systems, ICPS 2023, Wuhan, China, 8/05/23. https://doi.org/10.1109/ICPS58381.2023.10128049

Trajectory prediction method using deep learning for intelligent and connected vehicles. / Qie, Tianqi; Wang, Weida ; Yang, Chao et al.
Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023. Institute of Electrical and Electronics Engineers Inc., 2023. (Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Trajectory prediction method using deep learning for intelligent and connected vehicles

AU - Qie, Tianqi

AU - Wang, Weida

AU - Yang, Chao

AU - Li, Ying

AU - Zhang, Yuhang

AU - Liu, Wenjie

PY - 2023

Y1 - 2023

N2 - The trajectory prediction is significant for the driving safety of intelligent and connected vehicles. To accurately predict the vehicle trajectory, a hybrid method combining physic-based and data-based methods is proposed for intelligent and connected vehicles. The proposed method applied the physic-based method to represent vehicle kinematics. Then, the error of the physic-based method, which is the unmodeled features, is modeled with the data-based deep learning method using Encoder-Decoder Long short-term memory (LSTM). The proposed method is trained and evaluated by an actual vehicle dataset. When the prediction horizon is 3s, compared with the physic-based method, the longitudinal error, lateral error, and yaw angle error decreased by 93.9%, 86.6%, and 76.0%, respectively. Results show that the proposed method improves the trajectory prediction accuracy of autonomous and connected vehicles.

AB - The trajectory prediction is significant for the driving safety of intelligent and connected vehicles. To accurately predict the vehicle trajectory, a hybrid method combining physic-based and data-based methods is proposed for intelligent and connected vehicles. The proposed method applied the physic-based method to represent vehicle kinematics. Then, the error of the physic-based method, which is the unmodeled features, is modeled with the data-based deep learning method using Encoder-Decoder Long short-term memory (LSTM). The proposed method is trained and evaluated by an actual vehicle dataset. When the prediction horizon is 3s, compared with the physic-based method, the longitudinal error, lateral error, and yaw angle error decreased by 93.9%, 86.6%, and 76.0%, respectively. Results show that the proposed method improves the trajectory prediction accuracy of autonomous and connected vehicles.

KW - intelligent and connected vehicles

KW - long short-term memory (LSTM)

KW - trajectory prediction

UR - http://www.scopus.com/inward/record.url?scp=85163094436&partnerID=8YFLogxK

U2 - 10.1109/ICPS58381.2023.10128049

DO - 10.1109/ICPS58381.2023.10128049

M3 - Conference contribution

AN - SCOPUS:85163094436

T3 - Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023

BT - Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 6th IEEE International Conference on Industrial Cyber-Physical Systems, ICPS 2023

Y2 - 8 May 2023 through 11 May 2023

ER -

Qie T, Wang W , Yang C , Li Y, Zhang Y, Liu W. Trajectory prediction method using deep learning for intelligent and connected vehicles. In Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023. Institute of Electrical and Electronics Engineers Inc. 2023. (Proceedings - 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems, ICPS 2023). doi: 10.1109/ICPS58381.2023.10128049

Trajectory prediction method using deep learning for intelligent and connected vehicles

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this