ePrints

How does lane status influence drivers’ lane-change decisions? — An analysis based on naturalistic driving data

Lookup NU author(s): Dr Duo Li

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.

Abstract

Copyright © 2025. Published by Elsevier Ltd.Drivers must make lane-change decisions while interacting with surrounding vehicles in complex traffic environments. These decisions are highly sensitive to lane risk status, which leads to different behavioural outcomes. However, existing studies have not systematically explored the motivations and mechanisms of lane-changing behaviour under varying lane conditions. This study proposes a novel analytical framework using the highD dataset to examine how lane risk status affects drivers’ decision-making. By cross-classifying the status of the original and target lanes, four typical lane-change scenarios are identified. A two-stage method combining Association Rule Learning (ARL) and Structural Equation Modelling (SEM) is developed to extract behavioural patterns and establish interpretable causal mechanisms. Results show that drivers adopt distinct decision strategies under different lane-risk combinations. When the original lane is safe, they tend to prioritise greater headway for comfort, but excessive speed, neglect of speed differences, and the avoidance of large vehicles can increase the likelihood of risky lane changes. Conversely, when the original lane is risky, hazard avoidance and distance advantages become dominant, while vehicle-type preferences diminish. Moreover, the status of the target lane shifts drivers’ decision priorities. A safe target lane promotes the use of speed advantages to escape danger, while a risky target lane makes vehicle type the primary concern. These findings enhance the understanding of context-adaptive driving behaviour and provide theoretical support for intelligent lane-change strategies in autonomous driving systems.

Publication metadata

Author(s): Zhang H, Wang Y, Li D, Li J, Cao Y, Cheng Y, Ranjitkar P

Publication type: Article

Publication status: Published

Journal: Transportation Research Part F: Traffic Psychology and Behaviour

Year: 2026

Volume: 117

Print publication date: 01/02/2026

Online publication date: 22/11/2025

Acceptance date: 17/11/2025

ISSN (print): 1369-8478

ISSN (electronic): 1873-5517

Publisher: Elsevier Ltd

URL: https://doi.org/10.1016/j.trf.2025.103453

DOI: 10.1016/j.trf.2025.103453

Altmetrics

Share