Summary: PROLOGUE WP 3 - Video-based feedback for learner and novice drivers

The Austrian field trial of the PROLOGUE project tried to assess the impact of a videobased feedback system used during driver training on learner and novice drivers by analysing safety relevant behaviour from data collected in naturalistic circumstances.

Three types of driving events categorised in severity levels have been classified and risk scales were developed. Some significant differences on the risk scales occurred between subjects who received feedback and who did not: the feedback group showed less risk events after receiving their drivers licence compared to the control group.

  • Video feedback
Scope

The project "Promoting real Life Observations for Gaining Understanding of road user behaviour in Europe" - short PROLOGUE (Seventh Framework Programme of the European Commission) - sought to assess Naturalistic Driving observations by means of five independent field trials in order to show the current technical, methodological and organisational potential of this method. The project was led by the Dutch Institute of Road Safety Resarch - SWOV. The Austrian field trial tried to assess the impact of a videobased feedback system on learner and novice drivers by analysing safety relevant behaviour from data collected in naturalistic circumstances.

Method

Relevant driving parameters of 12 novice drivers were collected up to two months after receiving their driving licence. A data acquisition system was used as in-vehicle data recorder, which captured video signals, GPS position as well as acceleration forces. The novice drivers were divided into an intervention group (n=6), which received video-based feedback additional to the standard driving education and a control group (n=6), which got no such additional video-based feedback.

Three types of driving events categorised in severity levels have been classified, based on thresholds of either longitudinal or lateral acceleration values and a combination of both. Based on such events, risk scales were developed which are a measure of safety relevant events per hour driven.

Results

Some significant differences on the risk scales occurred between the intervention and control group showing more risk events for the control group after receiving their drivers licence compared to the intervention group. However, improvements concerning data reliability and automatic analyses procedures are necessary and the sample size of a large scale study would be needed for more reliable statistical results.