At higher levels of driving automation, drivers can nap during parts of the trip but must take over control in others. Awakening from a nap is marked by sleep inertia which is tackled by the NASA nap paradigm in aviation: Strategic on-flight naps are restricted to 40 min to avoid deep sleep and therefore sleep inertia. For future automated driving, there are currently no such strategies for addressing sleep inertia. Given the disparate requirements, it is uncertain whether the strategies derived from aviation can be readily applied to automated driving. Therefore, our study aimed to compare the effects of restricting the duration of nap opportunities following the NASA nap paradigm to the effects of sleep architecture on sleep inertia in takeover scenarios in automated driving. In our driving simulator study, 24 participants were invited to sleep during three automated drives. They were awakened after 20, 40, or 60 min and asked to manually complete an urban drive. We assessed how napping duration, last sleep stage before takeover, and varying proportions of light, stable, and deep sleep influenced self-reported sleepiness, takeover times, and the number of driving errors. Takeover times increased with nap duration, but sleepiness and driving errors did not. Instead, all measures were significantly influenced by sleep architecture. Sleepiness increased after awakening from light and stable sleep, and takeover times after awakening from light sleep. Takeover times also increased with higher proportions of stable sleep. The number of driving errors was significantly increased with the proportion of deep sleep and after awakenings from stable and deep sleep. These results suggest that sleep architecture, not nap duration, is crucial for predicting sleep inertia. Therefore, the NASA nap paradigm is not suitable for driving contexts. Future driver monitoring systems should assess the sleep architecture to predict and prevent sleep inertia.

The paper presents a study on motion sickness mitigation while driving with a fully automated vehicle on a test track. 31 participants who were susceptible to motion sickness experienced a 25-minuntes drive with multiple motion sickness provoking decelerations and accelerations while reading a text on a tablet. The participants experienced three different conditions in separate sessions: 1) control condition without countermeasure, 2) drive with active seat belt tensioner, 3) drive with a two-step driving profile. The participants rated their motion sickness on the MSTT Scale (during the drive) [1] and on the MSAQ (pre and post drive) [2]. After each drive, drivers rated their subjective experience of vehicle behavior and the countermeasures. On MSTT, the results showed no significant differences in the development of motion sickness across the three conditions. However, the two-step driving profile reduced the development of motion sickness assessed via MSAQ. Furthermore, both countermeasures seem to have the potential to positively influence the perception of the automation as safer, more trustworthy and more reliable.

Complex driving automation, the conditional availability of multiple levels of automation within the same vehicle, can challenge users in different situations. It therefore seems important to develop appropriate learning concepts that enable safe use. The objective of our work was to carry out a comprehensive analysis of user needs for such concepts.We conducted two complementary studies using three operational use cases as experimental manipulations: vehicle purchase, car-sharing rental, and over-the-air updates. First, an online study (N=242) examined current learning strategies, the relevance of learning opportunities, and user needs regarding instructional approaches, content, timing, and concept features. A subsequent interview study (N=21) explored the underlying motives across the three groups the motives and reasons behind their responses using a semi-structured interview guide.Drivers currently learn new functions primarily through trial-and-error. Learning concepts and most proposed contents were rated as relevant. Participants favored time concepts involving at least some pre-trip learning, considering 5–15 minutes acceptable. Video-based training, in-car tutors, simulator-based learning, and human instruction received high ratings, whereas manuals did not. Features supporting user autonomy and minimizing disruptions were perceived as crucial. No significant differences emerged across the three use cases.These findings identify key user needs for developing effective learning concepts for complex driving automation and highlight essential avenues for future research.

In automated driving, drivers may sleep during rides, making it important to study napping preferences and wake-up scenarios for human-centered system design. We conducted an online study with 280 participants, balanced by age and gender, examining how often, when, and how long individuals would nap if given the opportunity. The study also explored preferred wake-up methods when the vehicle nears the end of its operational design domain. Using a mixed 2 × 2 design, participants were assigned to one of two travel purposes (“commuting” vs. “holiday trip”) and two minimal risk conditions (rest area vs. hard shoulder). The results showed that 40% intended to sleep during automated driving, highlighting a strong interest in in-car sleeping. Wake-up preferences varied by travel purpose and minimal risk condition, with many favoring awakening before the drive ended. Participants also anticipated sleep inertia and desired post-wake-up support. The findings allow suggestions for designing wake-up concepts that mitigate sleep inertia.