Closing the Acceptance Gap: In-Car Training for Drivers to Utilise Assistance Systems
Repository URI
Repository DOI
Change log
Authors
Abstract
Road traffic accidents place a humanitarian and financial burden on society. Advanced Driver Assistance Systems (ADAS) can mitigate this as they increase driving safety. However, drivers’ acceptance rates are low and thus contradict drivers’ high appreciation of ADAS’ safety-enhancing effects, revealing a gap and impeding ADAS’ benefits. While several measures to close this gap have been suggested, no measure has fully achieved this.
The primary objective of this research was to establish an appropriate measure to close the aforementioned gap. Here, a mixed-methods approach embedded in the Design Research Methodology was used. First, a literature review on technology acceptance delivered numerous factors influencing ADAS acceptance without clarifying their precise contribution. Second, an online survey and a focus group determined the key factors causing low ADAS acceptance, i.e. awareness and mental models, and pointed to training as a potential measure to manipulate these factors. Third, a literature review on ADAS training highlighted an inconsistency of ADAS training’s effect on technology acceptance and added experience as another key factor. It also revealed the shortcomings of pre-usage training materials, i.e. neglecting awareness and experience. Fourth, a simulator study provided qualitative data on the effect of ADAS training on mental model accuracy. Here, it deduced that once an adequate mental model is reached a more accurate model does not increase ADAS acceptance further. Fifth, an on-road study indicated the possibility of dynamic in-car training by identifying factors causing high workload and showing the attainability of a workload prediction system. Sixth, an online survey revealed drivers’ appreciation of in-car training and their division over dynamic training. The collected data on drivers’ preferences was used to develop a high-level concept of an in-car training system. Finally, expert interviews refined this concept by determining dynamic training to be driver-initiated.
This thesis is the first research report to describe a concept of an in-car training system for ADAS and to present a comprehensive set of design guidelines for ADAS training. The suggested concept gives drivers control of their dynamic training, provides static training with a focus on ADAS’ benefits and raises awareness through targeted notifications. The guidelines advise on the training’s content, modality, and interaction design. They also recommend the usage of multiple technologies, e.g. smartphones and cars.