SAFETYLIT WEEKLY UPDATE

We compile citations and summaries of about 400 new articles every week.
RSS Feed

HELP: Tutorials | FAQ
CONTACT US: Contact info

Search Results

Journal Article

Citation

Zhang X, He S, Fan Z, Liu Y, Qi Y, Qu Y, Gu M, Niu Y, Fu W. Lect. Notes Elec. Eng. 2024; 1177: 267-276.

Copyright

(Copyright © 2024, Holtzbrinck Springer Nature Publishing Group)

DOI

10.1007/978-981-97-1103-1_24

PMID

unavailable

Abstract

Proceedings of 3rd 2023 International Conference on Autonomous Unmanned Systems (3rd ICAUS 2023)

It is such a challenge to ensure a safety driving when the self-driving vehicles are in a mixed traffic environment with human-driven vehicles. There are some abnormal driving behaviors, which could lead to accidents, financial losses, and even casualties. Therefore, the identification of abnormal driving behaviors is vital for the safety of autonomous vehicles, which could help to avoid the potential dangers. Considering that the abnormal behaviors take a small part of driving data, this paper proposed a Siamese neural network recognition model, which could be built up with small datasets. In detail, the vehicle trajectory data were preprocessed and used to train and validate the proposed method. Experimental results indicated that the Siamese neural network-based method could effectively identify abnormal driving behaviors with an accuracy rate of 88.0%. To some extent, the proposed method could assist the autonomous vehicles to recognize the abnormal driving behaviors successfully with a small sample size of driving data.


Language: en

NEW SEARCH


All SafetyLit records are available for automatic download to Zotero & Mendeley
Print