| Popis: |
Driver distraction is the diversion of attention to a non-driving related activity. It has been identified as major cause of accidents. Even as we move away from traditional ‘driver’ and towards highly-automated vehicles, distraction remains an important issue. A distracted driver could still potentially miss a handover of control message from the car, or have a reduced awareness of the traffic environment. With the increased number and complexity of new features being introduced in vehicles, it is becoming more important to understand how drivers interact with them, to understand the benefit they offer in helping the driver to focus on-road, but also to identify their limitations and risks. Thereby it is important to consider that the interaction between human and technology, e.g. driver distraction, can be described by many aspects. To learn the most about the interaction between user and technology, it is important to select a suitable measure and to utilise that measure in best practice, which can be hard to find in literature. This research project is divided into two research streams that investigate the opportunities of new in-vehicle interfaces to mitigate driver distraction and that research how to efficiently identify measures for the ergonomic evaluation of in-vehicle interfaces.\ud \ud Research stream one, comprising four studies, evaluated tactile information as a new interface technology to mitigate distraction in manual and automated cars. Tactile perception requires physical contact between the driver and the device delivering the feedback. It can be decreased by clothing. In the first user trial it was evaluated, for the first time, how shoe type, gender, and age influence the driver’s perception of a tactile pedal. Shoe type did not, but gender, age, and the feedback’s duration and amplitude did influence the perception. In some durations and amplitudes, the feedback was recognised by all participants and was rated highly intense, both aspects a warning should have. Next, it was evaluated how fast people would react to a tactile warning compared to a traditional auditory warning and an auditory-tactile warning. The participants reacted significantly slower to the tactile warning. Following, a tactile warning might not be suitable as an in-vehicle warning. However, adding an auditory component to the tactile warning increases its efficiency and people missed less auditory-tactile compared to auditory warnings. Newly introduced interfaces, such as tactile interfaces, put an effort on drivers to adjust to them and might lead to unsafe interactions. In the third and fourth study, it was investigated how a driver’s trust effects the reaction time and glance behaviour. Trust was not associated with the reaction time towards a tactile warning signal, but it influenced the glances at a voice-navigation interface that was new for the majority of the participants. The findings can be utilised to increase the trust in the interface dialogue and thereby decrease a driver’s time glanced off-road.\ud \ud Research stream two investigated how Human-Machine-Interface (HMI) engineers can be supported in the comparison and selection of measures (e.g. a usability score) to evaluate the ergonomics of in-vehicle devices, for example to measure driver distraction. Industry projects are often restricted by tight deadlines and limited availability of equipment. Measure selection can then become a time critical issue. In published literature, there existed no guidelines to support this task. In four rapid prototyping evaluations, an interface was developed that can aid HMI-engineers in the comparison and selection of measures for an ergonomic evaluation. The tool functions as knowledge management and foresees to inform users about the best practice to utilise a measure, tips to set-up required equipment, and templates for the measure, for example templates for the analysis or electronic versions of questionnaires. |
