How does a sudden request to take over manual driving affect the physiological activity (arousal) of users of automated vehicles?

A new publication by the WIVW on sleeping during an automated trip shows that drivers who are awakened by a takeover request exhibit a physiological reaction that seems paradoxical at first glance.

After awakening, patterns that are typical of sleep (e.g., slowed brain activity) coexist with patterns of increased activity (e.g., a sharp rise in heart rate). The findings could indicate that drivers need time to wake up completely after being awakened, but at the same time, they are also put under stress by the takeover request.

One explanation for these findings is that sleep inertia (a feeling of grogginess upon waking) causes stress in situations where drivers suddenly have to react to the traffic situation.

The study was published in the journal Nature Scientific Reports and is freely available here. It was conducted as part of the RUMBA project, which was funded by the German Federal Ministry for Economic Affairs and Climate Action (recently BMWE).

As part of the Hi-Drive User Training Programme, WIVW has developed a simulator-based training approach for Advanced Driver Assistance Systems (ADAS).

This format uses a driving simulator to allow users to engage with critical traffic scenarios, system boundaries and failure cases in a safe, controlled, and repeatable environment. The training combines a brief theoretical introduction with practical simulator sessions, allowing participants to familiarize themselves with the purpose and handling of the system. It complements the road-based formats offered by the mobility clubs involved in Hi-Drive.

A video about the training course can be found here.

Further information on the use of driving simulation for training and education can be found on the SILAB Driver Education & Training page.

How much situation awareness (SA) is really needed during automated driving (AD) to safely take over? How can we reliably measure situation awareness?

These are the questions we explored in our latest study at WIVW, which was conducted as part of the European Hi-Drive project and has now been published in Accident Analysis & Prevention. In a driving simulator experiment involving 41 participants, we manipulated visual input during AD to create four distinct levels of attention. Various methods were employed to assess SA, including subjective ratings, eye-tracking data, performance metrics and probe-based measures.

While subjective SA, gaze behaviour and performance varied across attention levels, complete visual suppression during automated driving had surprisingly little impact on takeover performance as long as the scenarios did not place high demands on SA. However, in more demanding situations, SA clearly played a critical role in ensuring good performance.

These findings offer important insights for evaluating and designing SA assessment methods in the context of automated driving.

You can access the full publication here for free.