Lane change warning threshold based on driver perception characteristics

Chang Wang; Qinyu Sun; Rui Fu; Zhen Li; Qiong Zhang

doi:10.1016/j.aap.2018.04.013

Lane change warning threshold based on driver perception characteristics

Accid Anal Prev. 2018 Aug:117:164-174. doi: 10.1016/j.aap.2018.04.013. Epub 2018 Apr 25.

Authors

Chang Wang¹, Qinyu Sun², Rui Fu³, Zhen Li⁴, Qiong Zhang⁵

Affiliations

¹ Chang'an University, School of Automobile, Middle Section of Nan'er Huan Road, Xi'an, 710064, China. Electronic address: wangchang@chd.edu.cn.
² Chang'an University, School of Automobile, Middle Section of Nan'er Huan Road, Xi'an, 710064, China. Electronic address: sunqinyu@chd.edu.cn.
³ Chang'an University, School of Automobile, Middle Section of Nan'er Huan Road, Xi'an, 710064, China. Electronic address: furui@chd.edu.cn.
⁴ Chang'an University, School of Automobile, Middle Section of Nan'er Huan Road, Xi'an, 710064, China. Electronic address: lizhen@chd.edu.cn.
⁵ Chang'an University, School of Automobile, Middle Section of Nan'er Huan Road, Xi'an, 710064, China. Electronic address: mingfangzhang@chd.edu.cn.

PMID: 29704793
DOI: 10.1016/j.aap.2018.04.013

Abstract

Lane Change Warning system (LCW) is exploited to alleviate driver workload and improve the safety performance of lane changes. Depending on the secure threshold, the lane change warning system could transmit caution to drivers. Although the system possesses substantial benefits, it may perturb the conventional operating of the driver and affect driver judgment if the warning threshold does not conform to the driver perception of safety. Therefore, it is essential to establish an appropriate warning threshold to enhance the accuracy rate and acceptability of the lane change warning system. This research aims to identify the threshold that conforms to the driver perception of the ability to safely change lanes with a rear vehicle fast approaching. We propose a theoretical warning model of lane change based on a safe minimum distance and deceleration of the rear vehicle. For the purpose of acquiring the different safety levels of lane changes, 30 licensed drivers are recruited and we obtain the extreme moments represented by driver perception characteristics from a Front Extremity Test and a Rear Extremity Test implemented on the freeway. The required deceleration of the rear vehicle corresponding to the extreme time is calculated according to the proposed model. In light of discrepancies in the deceleration in these extremity experiments, we determine two levels of a hierarchical warning system. The purpose of the primary warning is to remind drivers of the existence of potentially dangerous vehicles and the second warning is used to warn the driver to stop changing lanes immediately. We use the signal detection theory to analyze the data. Ultimately, we confirm that the first deceleration threshold is 1.5 m/s² and the second deceleration threshold is 2.7 m/s². The findings provide the basis for the algorithm design of LCW and enhance the acceptability of the intelligent system.

Keywords: Driver perception; Lane change warning; Signal detection theory; Threshold.

MeSH terms

Accidents, Traffic / prevention & control
Adult
Algorithms
Automobile Driving*
Deceleration*
Female
Humans
Male
Middle Aged
Models, Theoretical
Perception
Protective Devices / standards*
Risk Assessment