Lane change safety assessment of coaches in naturalistic driving state

•In this paper, we propose a new lane change safety identification model for coaches.•Lane change data for coaches are collected with a dynamic observation method.•A lane change safety model is built and examined using collected naturalistic data. The lane change behavior of coaches is more signific...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Safety science 2019-11, Vol.119, p.126-132
Hauptverfasser: Wang, Chang, Li, Zhen, Fu, Rui, Zhang, Mingfang, Sun, Qinyu
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•In this paper, we propose a new lane change safety identification model for coaches.•Lane change data for coaches are collected with a dynamic observation method.•A lane change safety model is built and examined using collected naturalistic data. The lane change behavior of coaches is more significant than other vehicles. We propose a new lane change safety identification model for coaches by considering the following behaviors during lane change. Naturalistic lane change data for coaches were collected by several sensors in an instrumented vehicle using a dynamic observation method. After discussing the lane change characteristics of coaches, a lane change safety model was established and examined using collected naturalistic data. The results showed that: (1) the average lane change duration was approximately 10.0 s, which is significantly higher than that of a car, 6.3 s; (2) coach drivers anticipate that lateral velocity during a lane change is stable at different velocities; and (3) lane change is similar to vehicle following behavior, considering longitudinal motion. Based on the lane change safety judge model proposed by Hussein Jula, 0.6 s time-headway (THW) was used to represent the longitudinal control behavior of coach drivers in the lane change safety model. Compared with ISO 17387 and other criteria, our model has significant advantages in the accurate identification rate of lane change safety, achieving 92.9%, 92.2% and 98.2% with 3 types of surrounding vehicles.
ISSN:0925-7535
1879-1042
DOI:10.1016/j.ssci.2018.09.009