Impact of in-vehicle navigation information on lane-change behavior in urban expressway diverge segments

• Published in: Accident analysis and prevention 2017-09, Vol.106, p.53-66
• Main Authors: Yun, Meiping, Zhao, Jing, Zhao, Jianzhen, Weng, Xuyan, Yang, Xiaoguang
• Format: Article
• Language: English
• Subjects: Accidents, Traffic - prevention & control; Adult; Automobile Driving - psychology; Computer Simulation; Deceleration; Diverge segments; Driving simulator; Female; Geographic Information Systems - statistics & numerical data; Humans; In-vehicle navigation; Lane-changing behavior; Male; Middle Aged
• ISSN: 0001-4575; 1879-2057
• DOI: 10.1016/j.aap.2017.05.025

•We examine the impacts of in-vehicle navigation information by driving simulator.•We test four information provision design schemes under three traffic flow status.•Lane-changing characteristics are analyzed by selecting six indicators.•In-vehicle navigation has positive impact on safety under medium/high density.•More improvement could be gained when the in-vehicle navigation provided earlier. Lane-changing behaviors frequently occur at upstream of the urban expressway off-ramp, which possibly cause bottleneck or accident. This paper studies how in-vehicle navigation information impacts lane-changing behavior at this urban expressway diverge segments with comparison to the traditional road signs by conducting a series of driving simulator experiments, in consideration of variation of the first provision time point and traffic flow density. Firstly, the driving simulator is validated by comparing data of lane-changing position and merging gap from both field survey and simulator. Then comparison study was performed for twelve scenarios which comprise four information provision design schemes under three different traffic flow density status. Lane-changing characteristics are analyzed by selecting six indicators, which are lane-changing merging gap, lane-changing position, lane change delay, lane-changing steering angle, lane-changing deceleration, and the safe distance of lane-changing. The results show that the impact of in-vehicle navigation information on lane-changing behavior varies with traffic flow density and the time point of the first navigation information provided. The in-vehicle navigation information had significant positive impact on lane-changing safety under medium to high density condition. However, the effect is not significant under light density condition. Moreover, more improvement in operational safety and smooth could be gained when the in-vehicle navigation information is provided earlier within range of 2km upstream of the exit gore.