Conference PaperPDF Available

Sleeping during highly automated driving - Target groups and relevant use cases of an in-car sleeping function

Authors:
  • Wuerzburg Institute for Traffic Sciences (WIVW GmbH)
  • Würzburg Institute for Traffic Sciences (WIVW GmbH)

Abstract and Figures

In highly automated driving (SAE Level 4), the driver will be no longer responsible for driving and can sleep during the ride. This opportunity is likely to change user needs. Our work focuses on the first phase of the user-centred design approach and aims to identify the target groups who are willing to use the sleep function and the relevant use cases. First, we conducted an online survey with N = 264 participants to investigate the characteristics that describe the future users of the sleep function. To derive relevant use cases, N = 7 participants of the online survey with a high intention to sleep during automated driving were invited to a subsequent interview study. The online survey identified predictors for a high intention to use a sleep function, such as young age as well as a high frequency and duration of sleeping as a passenger in public transport or cars. However, the results showed that there is no distinct target group. The interviews revealed that the wish to sleep during automated driving is related to the individual’s current mobility behaviour and the personal desire to enhance comfort during inconvenient trips. We derived exemplary use cases. Future research should identify requirements for comfortable sleep during highly automated driving.
Content may be subject to copyright.
In D. de Waard, S.H. Fairclough, K.A. Brookhuis, D. Manzey, L. Onnasch, A. Naumann, R. Wiczorek, F.
Di Nocera, S. Röttger, and A. Toffetti (Eds.) (2022). Proceedings of the Human Factors and Ergonomics
Society Europe Chapter 2022 Annual Conference. ISSN 2333-4959 (online). Available from http://hfes-
europe.org
Sleeping during highly automated driving target
groups and relevant use cases of an in-car sleeping
function
Markus Tomzig & Christina Kaß
Würzburg Institute for Traffic Sciences (WIVW)
Germany
Abstract
In highly automated driving (SAE Level 4), the driver will be no longer responsible
for driving and can sleep during the ride. This opportunity is likely to change user
needs. Our work focuses on the first phase of the user-centred design approach and
aims to identify the target groups who are willing to use the sleep function and the
relevant use cases. First, we conducted an online survey with N = 264 participants to
investigate the characteristics that describe the future users of the sleep function. To
derive relevant use cases, N = 7 participants of the online survey with a high intention
to sleep during automated driving were invited to a subsequent interview study. The
online survey identified predictors for a high intention to use a sleep function, such as
young age as well as a high frequency and duration of sleeping as a passenger in public
transport or cars. However, the results showed that there is no distinct target group.
The interviews revealed that the wish to sleep during automated driving is related to
the individual’s current mobility behaviour and the personal desire to enhance comfort
during inconvenient trips. We derived exemplary use cases. Future research should
identify requirements for comfortable sleep during highly automated driving.
Introduction
Nowadays, sleeping as a car driver is still a vision of future mobility. Previous studies
found relaxing, napping and sleeping as very popular non-driving related task (Becker
et al., 2018; Kyriakidis, Happee & de Winter, 2015). Whereas in today’s cars it is
neither possible nor legal to sleep as a car driver, this scenario could become real with
the advancing automation of vehicles. The Society of Automotive Engineers (SAE)
differs between six levels of automation (SAE, 2021). While in Levels 0 to 3, the
driver has to be at least ready to take over the driving task, in Levels 4 and 5 users do
not need to supervise the driving task. Sleeping as a driver would, thus, be legitimate
initially in Level 4, called “high driving automation” (SAE, 2021).
Up to now, it is unclear how such an in-car sleeping function must be designed to fit
user requirements. Before specifying user’s requirements and designing and realising
solutions, the user-centred design approach demands to specify the context of use
(DIN, 2010). Therefore, it is important to specify the target group and potential use
2 Tomzig & Kaß
cases of the function to be developed. The use cases do not only depend on the
technical feasibility but also on the requirements of the target group.
To identify the target group of a Level 4 sleeping function, we conducted an online
survey. The goal of this survey was to exploratively reveal characteristics which are
associated with the intention to sleep during highly automated driving and to outline
potential use cases. As a survey cannot explain the personal motives contributing to
the wish to sleep during automated driving, a subsequent interview study aimed to
gain a deeper understanding about the target group’s motives and needs that lead to
their wish to sleep in the car. The interviews ought to reveal distinct and attractive
usage scenarios that meet the needs of the target group.
Methods
Online survey
Sample and procedure
The participants were recruited from the test driver panel of the Wuerzburg Institute
for Traffic Sciences (WIVW), via project partners and social media. In total, N = 264
(n = 79 female, n = 185 male, n = 0 diverse) aged between 18 and 83 years (M = 47.1,
SD = 15.7) took part in the survey.
The survey was conducted in German language with the software LimeSurvey
(LimeSurvey GmbH). Completing the survey took about ten minutes. At the
beginning, the participants were informed that their participation was voluntary, not
financially compensated and that personal data would only be collected if the
participant was interested in taking part in the subsequent interview study. At the end,
participants could optionally provide their name and email to be invited in the
interview study. All data was collected and processed in accordance to the European
General Data Protection Regulation.
The survey was conducted between December 2020 and January 2021. At this time,
it had to be assumed that the Covid-19 pandemic significantly affected the mobility
of large parts of the society. For this reason, participants were asked to answer all
questions about their mobility as they had experienced it prior to the onset of the
pandemic, i.e., during a period with fewer or no home office and travel restrictions.
Measures
The main goal of the survey was the identification of characteristics describing a target
group for an in-car sleeping function. The criterion determining the target group was
defined by the intention to use such a future sleeping function. The intention to use
was measured by the level of agreement with the statement “In the future, I would
sleep during the automated drive if I had the opportunity to do so”. The approval was
indicated on a seven-point Likert-scale from “fully disagree” to “fully agree”.
To statistically describe the target group, several potential predictors have been
included in the survey. The investigated predictors addressed the following aspects:
socio-demographic characteristics (age, gender, educational level, amount of weekly
working hours, urbanisation of the residential environment), mobility (frequency of
target groups and relevant use cases of an in-car sleeping function 3
drives in the city, on rural roads, on motorways, at night, and in traffic jams as well
as the annual mileage, commuting time, the frequency of private and business trips),
sleeping behaviour (frequency and duration of naps as passenger in cars, in means of
public transport, and at home, as well as the feeling to get sufficient sleep in everyday
life), attitude towards driving (association of driving a car with fun, strain, stress, and
discomfort as well as the prior knowledge about automated driving).
The second goal of the survey was to explore the context, in which a sleeping function
is likely to be used by members of the target group. Therefore, all subjects who had
indicated at least slight agreement to the criterion were conceived as members of the
target group and were asked additional questions about their desired use cases of an
in-car sleeping function. The participants were asked about their preferred time of day
to use a sleeping function. Furthermore, they were asked about the estimated
frequency and duration of use. Additional items explored the desired occasion of the
drive and the types of road. A detailed list of all items and the used scales can be found
in table 5 in the appendix.
Data analysis
As there were no assumptions about a higher level model as well as a hierarchical
order of the predictors, all predictors were analysed by simple linear regressions.
Due to the large number of statistical tests, the p-values were adjusted according to
Bonferroni-Holm (Holm, 1979) within each category of predictor (socio-demographic
characteristics, mobility, sleeping behaviour, attitude towards driving). The level of
significance is α = .05 for all adjusted p-values.
Interview study
Sample and procedure
N = 7 participants (n = 3 female, n = 4 male, n = 0 diverse) aged between 32 and 63
years (M = 40.4, SD = 10.9) took part in the interview study. All participants had
previously participated in the online survey and indicated that they would use a future
in-car sleeping function. Therefore, they were conceived as members of the target
group (answer to the item “In the future, I would sleep during the automated drive if
I had the opportunity to do so” with at least “rather agree). On average, the sample
had a high intention to use (M = 6.4, SD = 0.8, Min = 5 “rather agree”, Max = 7 “fully
agree”).
After giving informed consent to the study procedure and data collection in
accordance to the European General Data Protection Regulation, the interview was
conducted as individual online video conference. The interview was conceptualised
as semi-structured, qualitative survey with an open answering form. Conducting the
interview took about 45 minutes.
Measures
The interview was structured in four thematic parts. The first part was dedicated to
better understand members of the target group. Participants were asked about the
reasons why they wished to sleep during highly automated driving, and which benefits
they saw in an in-car sleeping function. In the second block of questions, the
4 Tomzig & Kaß
participants were asked to describe in detail a desired use case of the in-car sleeping
function. The interviewer asked further questions, e.g., about the occasion,
destination, and duration of the drive, to what time of day it may take place and how
frequently these drives would occur. The third part was about the actual mobility of
the participants. The participants were asked how regularly and how frequently they
used a car and other means of transport (e.g., trains and busses) and also asked about
the trips’ occasions, durations and times of day. At the end of the interview, the
interviewer asked whether the participants’ mobility would change if there was the
possibility to sleep during the trip.
Data analysis
The participants’ qualitative verbal responses were tagged and clustered by content,
meaning that responses were grouped across different questions if the participant
provided related information herein. Because of the small sample size, inference
statistical tests are not indicated, and all analyses were carried out qualitatively and
descriptively.
Results
Online survey
Relationship between the investigated predictors and the intention to use an in-car
sleeping function
Of the N = 264 surveyed participants, n = 106 agreed at least slightly to the criterion
(n = 47 “rather agree”, n = 42 “agree”, n = 17 “fully agree”). The proportion of these
as target group conceived persons in the entire sample was thus 40.2%. Of the
remaining participants, n = 147 rejected the idea of sleeping during the drive (n = 44
“rather disagree”, n = 41 “disagree”, n = 57 “fully disagree”). The remaining n = 16
participants indicated “neither nor”.
Among the socio-demographic characteristics, age significantly predicted the
intention to use the sleeping function, indicating that younger participants had a higher
intention to use than older ones. Further, the educational level significantly predicted
the intention to use: A higher educational degree was associated with a higher
intention to use. The gender, amount of weekly working hours and urbanisation of the
residential environment could not predict the intention to use. The statistical results
of the socio-demographic variables can be found in table 1.
Table 1. Results of linear regressions for socio-demographic variables
Variable
β
Adj.R²
F
df
p
Adj. p
Age
-.035
.071
21.130
1, 261
< .001
< .001
Gender
.224
< .001
0.703
1, 262
.402
.402
Highest educational
level
.295
.085
25.450
1, 262
< .001
< .001
Amount of weekly
working hours
.008
< .001
1.022
1, 262
0.313
.626
Urbanisation of the
residential environment
.189
.011
3.934
1, 262
0.048
.144
target groups and relevant use cases of an in-car sleeping function 5
After Bonferroni-Holm adjustment, the participants’ mobility did not significantly
predict the intention to use. There were tendencies that the intention to use was
predicted by the frequency of night drives and the frequency of traffic jams, indicating
that people may have a lower intention to use if they experience frequent night drives
or traffic jams. However, after adjusting the p-values, these models are conceived as
not significant. The statistical results of these and the other mobility variables are
listed in table 2.
Table 2. Results of linear regressions for mobility variables
Variable
β
Adj.R²
F
df
p
Adj. p
Frequency of drives in
the city
-.162
.005
2.191
1, 262
.140
.840
Frequency of drives on
rural roads
-.171
.005
2.282
1, 262
.132
.924
Frequency of drives on
motorways
-.075
< .001
0.496
1, 262
.482
> .999
Frequency of drives at
night
-.247
.011
3.881
1, 262
.050
.398
Frequency of drives in
traffic jams
-.410
.016
5.220
1, 262
.023
.207
Annual mileage
-.038
< .001
0.123
1, 262
.726
> .999
Commuting time
(single way)
.089
< .001
0.550
1, 187
.459
> .999
Frequency of business
trips longer than 2
hours
.119
< .001
0.721
1, 262
.397
> .999
Frequency of private
trips longer than 2
hours
.057
< .001
0.115
1, 262
.734
.734
In the domain of sleeping behaviour, the frequency and duration of taking a nap as
passenger in cars in the present significantly predicted the intention to use. Passengers
who usually sleep longer and more often had a higher intention to use an in-car
sleeping function. The same applies for persons who sleep regularly and for long
periods in public means of transport. These variables also significantly predicted the
intention to use. In contrast, the frequency of taking a nap at home did not significantly
predict the wish to use a sleeping function. However, participants who perceived their
regular nocturnal sleep as insufficient had a higher intention to use than people with
sufficient sleep. The statistical results of all variables concerning sleeping behaviour
are listed in table 3.
6 Tomzig & Kaß
Table 3. Results of linear regressions for variables concerning sleeping behaviour
Variable
β
Adj. R²
F
df
p
Adj. p
Frequency of naps as
passenger in cars on
trips longer than 40
minutes
.663
.102
27.020
1 ,228
< .001
< .001
Duration of naps as
passenger in cars on
trips longer than 40
minutes
.630
.096
25.850
1, 233
< .001
< .001
Frequency of naps as
passenger in public
means of transport on
trips longer than 40
minutes
.802
.175
44.760
1, 205
< .001
< .001
Duration of naps as
passenger in public
means of transport on
trips longer than 40
minutes
.486
.121
29.270
1, 205
< .001
< .001
Frequency of naps at
home up to 40 minutes
.057
< .001
0.144
1, 259
.704
> .999
Feeling of getting
sufficient sleep
-.202
.021
6.517
1, 259
.011
.022
All measured variables about the attitude towards driving significantly predicted the
intention to use an in-car sleeping function. Participants who associated driving with
strain, stress, or discomfort and did not associate driving with fun had a higher
intention to use. Prior knowledge about automated driving correlated positively with
the appreciation of an in-car sleeping function. The statistical results about the attitude
towards driving are listed in table 4.
Table 4. Results of linear regressions for variables concerning attitudes towards driving
Variable
β
Adj. R²
F
df
p
Adj. p
Association of driving a
car with fun
-.241
.026
7.932
1, 262
.005
.021
Association of driving a
car with strain
.212
.029
8.868
1, 262
.003
.016
Association of driving a
car with stress
.216
.025
7.619
1, 262
.006
.012
Association of driving a
car with discomfort
.203
.015
5.130
1, 262
.024
.024
Prior knowledge about
automated driving
.166
.025
7.753
1, 262
.006
.017
target groups and relevant use cases of an in-car sleeping function 7
Characteristics of use cases for an in-car sleeping function
The n = 106 participants with an intention to use an in-car sleeping function were
asked the further questions about the circumstances in which they could imagine
sleeping during the drive. The majority of the subsample wanted to use the function
occasionally (n = 44) or rarely (n = 35). N = 14 imagined using the sleeping function
frequently, n = 4 very frequently. N = 9 would (almost) never sleep during automated
driving. The minimum imagined travel time to make use of the in-car sleeping
function was on average M = 77.5 minutes (SD = 76.3, min = 10, max = 480).
The majority of the target group could imagine sleeping on drives to their holiday
destination (n = 93) or on recreational trips (n = 63). For daily routine trips (e.g.,
shopping), a sleeping function does not seem to be a promising feature (n = 5). In the
domain of job-related occasions, n = 61 wanted to sleep on their way from work,
followed by n = 52 who wished to sleep on vocational drives, and n = 38 on their way
to work.
Sleeping during automated driving was conceivable at all times of day with a
preference of the early morning and the night (cf. figure 1). A sleeping function would
be used mainly on motorways (n = 95) and during traffic jams (n = 90). The vision of
sleeping on federal roads (n = 57) or straight rural roads (n = 55) found partial appeal.
Sleeping on urban roads (n = 17) or on bendy rural roads (n = 12) does not seem to be
promising.
Figure 1. Intended times of day to use an in-car sleeping function.
Interview study
Reasons for the wish to sleep during automated driving
When asked about the reasons for their intention to use an in-car sleeping function,
participants mentioned both, problems they associate with manual driving and
solutions they expect from an in-car sleeping function. With regard to today’s
8 Tomzig & Kaß
problems, participants indicated that longer trips often evoke strain, stress, and
persistent concentration. As a result, participants reported that they quickly become
tired during long or overnight car trips and need to plan their driving times according
to their fitness level. One participant added feeling uncomfortable before long trips.
Consistent with the aforementioned problems, on the benefits side, participants
reported that sleeping during the trip offered the opportunity to relax and recover. The
vision of an in-car sleeping function was further associated with higher driving
comfort. It was expected that this would make one arrive at the destination recovered
instead of fatigued. In addition to the aspect of comfort, the participants also
mentioned a gain in time due to the in-car sleeping function. The time of day could be
utilized better and a lack of sleep e.g., from the previous night, could be caught up
during the trip. Furthermore, it would be more attractive to travel at off-peak times
than it is nowadays.
Derived use cases of an in-car sleeping function
The scenarios described by the participants were clustered into three use cases. Most
participants named more than one imaginable use case. The majority (n = 6) wished
to sleep on drives to visit their families and/or friends who do not live nearby. The
described trips took between one and five hours and were typically made on weekends
(e.g., as weekend commuters). The mentioned drives began in the late afternoon and
ended in the early or late evening, depending on the driven distance. With regard to
the taken sleep, parts of the nocturnal sleep would be translocated into the vehicle.
N = 5 participants wished to sleep (also) on shorter trips that take between 30 and 45
minutes and start either in the early morning (outbound) or in the late afternoon
(return). As occasions, participants named destinations for one day, mostly to get to
and from work (n = 4) but also recreational trips (n = 1). In this use case, the
participants would rather nap instead of sleeping deeply.
Third, n = 3 participants described the wish to use an in-car sleeping function during
long drives to their holiday destination. The participants indicated that they already
partially do these trips at night today. Traveling time is supposed to be between 8 and
10 hours. With an in-car sleeping function, the participants anticipated to sleep
through the entire night and to arrive recovered.
Anticipated changes in mobility due to an in-car sleeping function
N = 3 participants indicated that their mobility behaviour would change due to the
possibility to sleep during the trip. In terms of changes, the participants named to make
the most of holiday trips, to travel long distances more often, and to generally travel
more often and at other times. N = 2 participants felt that their mobility would change
only slightly. They expected to have to plan less before trips to be fit to drive.
Furthermore, they wanted to reschedule their holiday journeys into the night. The
remaining n = 2 participants anticipated no changes in their mobility behaviour due to
an in-car sleeping function. Their driving routes and driving times were already fixed
and could not be changed easily (e.g., way to work).
target groups and relevant use cases of an in-car sleeping function 9
Discussion
The development of high driving automation will enable the driver to completely
refrain from the driving task and to sleep during the trip. The presented studies focused
on the first phase of the user-centred design approach and aimed at identifying the
target groups who are willing to use an in-car sleeping function and to reveal the
relevant use cases with an exploratory approach. An online survey with N = 264
participants and a subsequent interview study with N = 7 participants have been
conducted.
Implication of results
The results of the online survey demonstrate that an in-car sleeping function is
perceived as an attractive feature. The intention to use cannot be ascribed to a distinct
target group but is associated with several predictors such as young age, high
education, regular naps in means of transport, and aversion to manual driving.
The online survey examined the attractiveness of different scenarios of use (duration
and occasion of the journey, road conditions, and the time of day). Especially long
trips and trips to recreational and holiday destinations found agreement. Furthermore,
sleeping during automated driving is imaginable at any time of day, with focus on the
night and early morning. However, the online survey cannot explain the personal
motives contributing to the wish to sleep during automated driving. The survey’s
results reveal characteristics for future use cases, but they cannot explain how these
characteristics are related to one another.
Targeting on these issues, the results of the subsequent interview study showed that
members of the target group expect an eased and restorative instead of a stressful and
straining drive. Members of the target group are unified by the common wish to
improve recurring straining drives they are already experiencing today. The
interviews showed how the characteristics of the expected use cases matched to one
another: The participants named day trips taking between 30 and 45 minutes starting
in the early morning and/or late afternoon, weekend trips taking between one and five
hours starting in the late afternoon, and long holiday trips taking between eight and
ten hours. These results clarify that an in-car sleeping function must address a wide
application area. Members of the target group expect short naps as well as long, deep
sleep. This should be considered in the further ergonomic development of the sleeping
function. A sleeping function should be able to be used at all times of day. In contrast,
the interviews provided evidence that long sleeping periods during the day are as
unlikely as short powernaps during night trips.
A considerable part of the interview sample expected changes in their mobility due to
the availability of an in-car sleeping function. However, we consider the indicated
changes as rather small. Essentially, the participants expected to travel longer
distances and more often. Major changes in mobility, e.g., changes in their place of
residence or job have not been mentioned.
10 Tomzig & Kaß
Methodological limitations
The results are supposed to provide a forecast about future target groups and use cases
of an in-car sleeping function. The accuracy of this forecast may be limited by the fact
that the surveyed sample is not representative to the general population and may have
been subject to a self-selection. It is, thus, possible that the participants had a higher
interest to the survey’s topic than the general population which may lead to an
overestimated interest in the sleeping function. The actual proportion of the target
group by the entire population may, thus, be overestimated. Further, the interviews
examined only a very small subsample and gathered qualitative results and
perspectives of few persons. Conclusions to the general population must therefore be
drawn very carefully. However, we consider that the subsample fairly represents the
target group as the subsample’s results in the interviews fitted the results of the entire
sample in the online survey. Moreover, the interviews provided complementary
impressions about needs and motives of the target group.
The quality of the forecast may further be reduced due to the fact that to this point of
time, high driving automation is not generally available and sleeping as driver is not
possible yet. Therefore, the results base on today’s needs which can be solved by a
future in-car sleeping function. This may explain that the use cases reflect mobility
behaviour that is rather typical for today’s time. It must be considered that needs and
mobility may change with emerging new technologies. It cannot be excluded that high
driving automation and the possibility to sleep during the trip encourage new patterns
of mobility which can hardly be discovered by an early user-centred approach.
Likewise, up to this point, little is known about the technical conditions, possibilities,
and limitations of a sleeping function. It is therefore unclear whether the user
requirements are technically feasible (e.g., highly automated driving for eight hours
to holiday destination).
Conclusion
The online survey showed that there is not one distinct target group that would like to
use an in-car sleeping function, but that there are several indicators that predict the
intention to use. The interviews explored conceivable use cases. A sleeping function
is appreciated if recurring drives are perceived as inconvenient (e.g., associated with
fatigue). Further research is now required to examine the user requirements towards
the design of an in-car sleeping function. To present design solutions, technical
development must also be taken into account.
References
Becker, T., Herrmann, F., Duwe, D., Stegmüller, S., Röckle, F., & Unger, N. (2018).
Enabling the value of time. Implications for the interior design of
autonomous vehicles. Stuttgart, Germany: Fraunhofer IAO.
DIN. (2010). Ergonomics of human-system interaction - Part 210: Human-centred
design for interactive systems (ISO 9241-210:2019); German version.
Berlin, Germany: Deutsches Institut für Normung.
target groups and relevant use cases of an in-car sleeping function 11
Holm, S. (1979). A simple sequentially rejective multiple test procedure.
Scandinavian journal of statistics, 6, 65-70.
Kyriakidis, M., Happee, R., & de Winter, J.C. (2015). Public opinion on automated
driving: Results of an international questionnaire among 5000 respondents.
Transportation Research Part F: Traffic Psychology and Behaviour, 32,
127-140.
SAE. (2021). Taxonomy and Definitions for Terms Related to Driving Automation
Systems for On-Road Motor Vehicles (Vol. J3016). SAE International.
12 Tomzig & Kaß
Appendix
Table 5. Items used in the online survey
Category
Item
Socio-
demographics
1.
Age
2.
Gender
3.
Highest educational level
4.
Amount of weekly working
hours
5.
Urbanisation of the
residential environment
Mobility
6.
Frequency of drives in the
city
7.
Frequency of drives on
rural roads
8.
Frequency of drives on
motorways
9.
Frequency of drives at
night
10.
Frequency of drives in
traffic jams
11.
Annual mileage
target groups and relevant use cases of an in-car sleeping function 13
Category
Item
12.
Commuting time (single
way)
13.
Frequency of business trips
longer than 2 hours
14.
Frequency of private trips
longer than 2 hours
Sleeping
behaviour
15.
Frequency of naps as
passenger in cars on trips
longer than 40 minutes
16.
Duration of naps as
passenger in cars on trips
longer than 40 minutes
17.
Frequency of naps as
passenger in public means
of transport on trips longer
than 40 minutes
18.
Duration of naps as
passenger in public means
of transport on trips longer
than 40 minutes
19.
Frequency of naps at home
up to 40 minutes
20.
Feeling of getting sufficient
sleep
14 Tomzig & Kaß
Category
Item
Attitude
towards
driving
21.
Association of driving a car
with fun
22.
Association of driving a car
with strain
23.
Association of driving a car
with stress
24.
Association of driving a car
with discomfort
25.
Prior knowledge about
automated driving (self-
assessment)
Use case
26.
Time of day to use an in-
car sleeping function
27.
Minimum duration of trip
for using sleeping function
28.
Anticipated frequency of
use
29.
Acceptable road conditions
to use a sleeping function
30.
Anticipated driving
occasions for using
sleeping function
... Vehicles which are equipped with an automated driving system corresponding to SAE Level 4 or higher (SAE, 2021) may offer the driver the opportunity to take a nap during the trip. Napping during the automated drive is a very popular future use case (Becker et al., 2018;Tomzig & Kaß, 2022) and has the potential to restore self-reported mood and objective performance during subsequent manual driving (Hartzler, 2014;Milner & Cote, 2009;Tietzel & Lack, 2001). Some promote additional nap opportunities during automated driving even as solution against "America's sleep deprivation problem" (Eliot, 2019). ...
Article
Full-text available
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.
Article
This paper presents a simple and widely ap- plicable multiple test procedure of the sequentially rejective type, i.e. hypotheses are rejected one at a tine until no further rejections can be done. It is shown that the test has a prescribed level of significance protection against error of the first kind for any combination of true hypotheses. The power properties of the test and a number of possible applications are also discussed.
Ergonomics of human-system interaction -Part 210: Human-centred design for interactive systems
  • Din
DIN. (2010). Ergonomics of human-system interaction -Part 210: Human-centred design for interactive systems (ISO 9241-210:2019);
Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles
  • Sae
SAE. (2021). Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles (Vol. J3016). SAE International.