Application of global positioning system and questionnaires data for the study of driver behaviour on two‐lane rural roads

Abstract

ABSTRACT:Methodologies based on naturalistic observation provide the most accurate data for studying drivers' behaviour. This paper presents a new methodology to obtain naturalistic data related to drivers' behaviour in a road segment. It is based on the combination of using global positioning system data and drivers' questionnaires. The continuous speed profiles along a road segment and the characteristics of drivers, of their trips and the type of their vehicles can be obtained for a great amount of drivers. It has already been successfully used for several studies, such as the development of models to estimate operating speed profile in two-lane rural road segments; or the characterization of driving styles. These operating speed models have been the key for the development of a new geometric design consistency model, allowing an easier road safety evaluation. Besides, knowledge on the human factors that influence speed choice may be useful for road safety media campaigns and education programs designers, and also for the improvement of intelligent driver assistance systems. 2 INTRODUCTIONMultiple factors typically combine to produce circumstances that lead a vehicle to crash. The main concurrent factors are human factors, roadway environment factors, and vehicle factors. According to the research by Treat et al. [1], human factors are the most prevalent contributing factor of crashes, being present on the 93 % of crashes; followed by roadway (34 %) and vehicle factors (13 %).However, due to its characteristics, it is also the most complicated factor to study. The best results in this area have been achieved by using data collection methodologies based on naturalistic observation. This kind of methodologies is based on subjects driving the way they usually do, in their own vehicles and without any specific instructions or interventions. Whereas 100-Cars Naturalistic Driving Study was focused on drivers' behaviour at crash situations, the central goal of SHRP2 Naturalistic Driving Study is the study of how the driver interacts with and adapts to the vehicle, traffic environment, roadway characteristics, traffic control devices, and the environment. It also includes assessing the changes in collision risk associated with each of these factors and interactions.This research is involving more than 3000 volunteer drivers, who are monitored for a three year period, beginning in 2010.Both previous research mainly study passenger car drivers' behaviour. 2 Be Safe aims to understand and characterize powered two wheeler (PTW) rider behaviour. This is the world's first naturalistic riding study involving instrumented PTWs.In most of that kind of studies, drivers are volunteers who know the scope of the research project, so their behaviour may be biased. Besides, they mainly focus on drivers' behaviour at crash situations, which are rare events. For characterizing drivers' behaviour, the most studied variable is the speed at which they drive at different road alignment elements (curves, tangents and spiral transitions) a...

Estudio de la velocidad desarrollada por vehículos pesados en carreteras convencionales

“…Teniendo en cuenta que el control sobre el recorrido realizado por los mismos y sus condiciones de circulación fueron mínimas, es necesario estudiar cada uno de estos perfiles para garantizar que se incluyen en el estudio únicamente aquellos circulando en flujo libre. La metodología utilizada para la evaluación de las condiciones de circulación en flujo libre ha sido la desarrollada por Pérez-Zuriaga et al (2013). Está basada en la hipótesis de que cada conductor se comporta de una forma particular, aproximándose su perfil de velocidad individual al de un percentil.…”

Development and Calibration of a Global Geometric Design Consistency Model for Two-Lane Rural Highways, Based on the Use of Continuous Operating Speed Profiles

“…Those speed measurements were obtained by means 134 of video recording, far away from drivers' line of sight. Results showed that drivers were no biased by the presence of 135 GPS devices Pérez-Zuriaga et al 2013). 136 37 tangent-to-curve transitions were identified from those ten road segments.…”

“…This selection was carried out after 138 an in-depth analysis of the operating speed profiles for all drivers. Vehicles travelling at non-free-flow conditions were 139 also discarded from the analysis (Pérez-Zuriaga et al 2013). The number of vehicles at each tangent-to-curve 140 transition ranged from 24 to 102.…”

“…Even in the case they are not aware of them, they are driving an instrumented vehicle which is not their own vehicle. Those devices and the different vehicle per se make it difficult for drivers to perform naturally, thus affecting the results.An innovative continuous data collection methodology was presented by Pérez-Zuriaga et al (2013). This consists on giving small GPS tracking devices to actual users of a certain road.…”

“…An innovative continuous data collection methodology was presented by Pérez-Zuriaga et al (2013). This consists on giving small GPS tracking devices to actual users of a certain road.…”