Positioning Eye Fixation and Vehicle Movement: Visual-motor Coordination Assessment in Naturalistic Driving

Abstract

In recent years, many driving studies in the traffic safety literature have undertaken error assessments of driver behaviour. However, few studies have been able to analyse the detailed individual vision and motor behaviours of drivers, due to the lack of reliable data and available technologies. Therefore, little is currently known about drivers' visual-motor coordination involving the use of visual information to regulate their physical movements. This research sets-up a technical framework to investigate on-road drivers' visual-motor coordination via vision tracking and vehicle positioning. The driving behaviour and performance were recorded and analysed using Eye Movement Tracking, Global Navigation Satellite System (GNSS) and Geographic Information Systems (GIS). The eye tracker recorded eye fixations and duration on video images to analyse the visual pattern of individual drivers. Real-time kinematic (RTK) post-processing of multi-GNSS generated vehicle movement trajectory at centimetre-level accuracy horizontally, which encompasses precise lateral positioning, speed and acceleration parameters of driving behaviours. The eye fixation data was then geocoded and synchronised with the vehicle movement trajectory in order to investigate the visual-motor coordination of the drivers. A prototype of implementation of the framework focusing on complex U-turn manoeuvre at a roundabout in five older drivers was presented in this paper. The visualisation of spatial-temporal patterns of visual-motor coordination for individual drivers allows for a greater insight to behaviour assessment. The on-road driving test in this study has also demonstrated a discriminant and ecologically valid approach in driving behaviour assessment, which can be used in studies with other cohort populations