Human errors contribute to more than 90% of all traffic accidents. While automated driving systems (ADS) can potentially reduce human errors in the future, they also increase the complexity of driver-vehicle interactions. Therefore, optimising these interactions plays a critical role in improving road safety for the future.
The first part of this thesis used exploratory research methods to investigate the potential of attentive driver-vehicle interfaces to improve takeover performance in vehicles with conditionally automated driving systems (CADS). While these CADS (SAE Level 3) can handle the entire dynamic driving task under certain conditions, they require the driver to take control when the system fails or exceeds the operational design domain (ODD).
Therefore, the ability of the driver to effectively regain control of the CADS has been identified as a critical factor in improving the performance of the takeover. Since the driver’s level of SA is essential for a good takeover performance and starts to decrease after the ADS is activated, improving the SA to an optimal level is critical for enhancing takeover performance.
To address this issue in the second part, the author conducted an experimental driving simulator study using the activation strategy of attentive user interfaces to increase the driver’s level of arousal to an optimal level.
In this study, an attentive driver-vehicle interface presented participants with a trivia task during which they were asked questions about the driving environment via a voice user interface. While this study did not find significant evidence that the activation strategy can be used to improve takeover performance in conditionally automated vehicles, the data suggest that the trivia task positively influenced the driver’s level of SA in the experimental condition.
