Conferences and supporting programme
A driving simulator study investigating the effects of digital mirror failures on visual and driving behaviour, ...
In the concept of mirrorless cars, traditional side mirrors are replaced with camera-based displays within vehicles, thereby improving vehicle aerodynamics and the field of view. Digital technology may fail, however. Besides faults in the software and electronic circuits, camera-based systems are susceptible to environmental factors that may limit the camera’s vision, such as rain and ice, dirt, or image distortions due to extremely bright or low sunlight conditions. Specifically, for situations in which drivers' view of the sides and back of their car depends on digital mirrors, malfunctions may result in insufficient situation awareness and unsafe manoeuvres. Moreover, display problems may lead to substantial distraction, as drivers may attempt to extract information from a degraded source. The current study aimed to determine whether drivers were distracted by a digital mirror failure, reported increased mental workload and responded with compensatory behaviours. Effects on visual behaviours may involve increased rear-view mirror and over the shoulder checks, and drivers may reduce their speed. The research also aimed to investigate whether any changes lasted into future drives, when the mirror resumed correct functionality. An experiment was conducted using a UK motorway scenario in a medium-fidelity driving simulator at the University of Nottingham. For this study, the LCD wing mirrors of the simulator car were replaced with separate LCD panels inside the vehicle. In three drives (each approximately 10 minutes long), 30 regular drivers performed several lane-change manoeuvres while being surrounded by ambient traffic. In order to measure the effects of digital mirror failures, the drivers were subjected to a failure condition of the right (offside) digital mirror, at a dedicated but unpredictable time, immediately after being instructed to move from the middle into the right hand (fast) lane. This failure always occurred during Drive 2. Participants experienced either a blank (no information), a degraded (hard to extract information) and a frozen (misleading information) display. The analysis compared the lane change in which the failure occurred (Drive 2) with the corresponding lane changes in the drives without failures (Drives 1 and 3). Participants' reactions were recorded by the driving simulator software, cameras inside the vehicle and workload was measured with the NASA-TLX questionnaire. Results show that, in the lane change following the failure, the drivers compensated by longer and more glances towards the right mirror and a slight increase in over the shoulder checks. The drivers also increased speed and lateral variability, and reported heightened workload. There were indications that the blanked mirror affected visual distraction and workload less than the other failures. The blank image could have possibly communicated the failure clearly and also motivated the drivers to perform the compensatory behaviours. There were no significant differences between the first and third drives, which were free from failures, indicating no lasting effects of the failure. Future research needs to consider digital mirror failures in the real world with a wider range of different manoeuvres.
--- Date: 28.02.2019 Time: 11:20 - 11:40 Location: Conference Counter NCC Ost