Safer secondary tasks in partially 
self-driving vehicles 
Master’s thesis in Industrial Design Engineering 
 
 
 
CHENGLIN SONG 
RAN AN 
 
 
 
 
 
 
 
 
 
 
DEPARTMENT OF INDUSTRIAL AND MATERIALS SCIENCE 
DIVISION DESIGN & HUMAN FACTORS 

CHALMERS UNIVERSITY OF TECHNOLOGY 
Gothenburg, Sweden 2022 
www.chalmers.se 

 

http://www.chalmers.se/


 
 
 
 
 

 
 

  



 
 
 
 
 

 
 

 
MASTER’S THESIS 2022 

 
 
 

Safer secondary tasks in partially  
self-driving vehicles 

 
 

CHENGLIN SONG 
RAN AN 

 
 
 
 
 
 
 
 

 

 
 
 
 
 
 

Department of Industrial and Materials Science  
Division of Design & Human Factors  

CHALMERS UNIVERSITY OF TECHNOLOGY  
Gothenburg, Sweden 2022  

  



 
 
 
 
 

 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 
Safer secondary tasks in partially self-driving vehicles 
 
Master of Science Thesis  
 
In collaboration with Autoliv Sverige AB 
  
© Chenglin Song & Ran An, 2022 
 
Department of Industrial and Material Science  
Chalmers University of Technology  
SE-412 96 Göteborg, Sweden  
Phone +46(0) 31 - 772 10 00  
Cover illustration: Steering wheel screen concept, by Chenglin Song & Ran An  
 
Printed by Repro Service Chalmers  
Gothenburg, 2022   



Abstract
This Master Thesis was conducted at the Industrial Design Engineering program at
Chalmers University of Technology in collaboration with Autoliv Sverige AB.

Self-driving cars from Level 1 to Level 3 that are currently ready to be put into the
market, the proportion of time that the system controls the car gradually increases. In
the future, higher-level autonomous driving technology can further reduce the
distraction of perception factors, freeing people's hands to do other things, such as
checking emails, watching videos, and reading. We believe that there is a possibility
of a screen on the steering wheel of the car to perform secondary tasks.

The aim was to come up with conceptual graphical user interface design alternatives
for the touch screen control that could provide safer secondary tasks during SAE L3
automation.  Academic research in related fields, user study and benchmarking were
performed before designing. The deliverables were two concepts for steering wheel
screens in different directions, including concept sketches and high-fidelity
prototypes of the user interfaces for Gmail and Youtube respectively of the two
concepts.

Concept I uses a mobile phone as a screen mounted on the steering wheel to
present information and provide control functions. The phone could be connected
with the vehicle system to display the user interface through the center screen. In
addition the user can use a stick and a voice assistance physical button to control
the center screen for secondary tasks. Concept II integrates a touchpad screen on
the right-hand side of the steering wheel as a small display and controller for the
central display. The screen can switch between a display and a touchpad when
performing secondary tasks.

A final user testing was performed after building the high-fidelity prototypes. The test
results show that the proposed concepts can meet the project objectives to a certain
extent, and further professional testing is needed. This thesis provides a feasible
scheme for level 3 autonomous vehicles to perform secondary tasks, and open up
new possibilities for level 3 and other autonomous vehicles to perform secondary
tasks.



Acknowledge
Over the process of researching and writing this paper, we would like to express our
thanks to all those who have helped us.

First, we would like to express our gratitude to our examinar and supervisor, Bijan
Aryana, who gave us lots of suggestions during the process of researching.

Sincere gratitude should also go to Autoliv and our colleagues, especially our
manager Eric Moragues who gave us utmost help and encouragement for
researching and designing. It is an honor to work with Autoliv and Eric on the master
thesis project.

We would also like to thank all the participants in the user studies and user tests.
Thanks for their time and effort they invested in this project.

Finally, our gratitude should also go to Jinhong Guo and Yodit Kinfe, our opponents,
for your helpful feedback on our work.

Gothenburg, June 10th 2022
Ran An & Chenglin Song



Table of Contents

1 Introduction 1
1.1 Background 1
1.2 Aim and Scope 3

2 Literature Review 4
2.1 Aim 4
2.2 Findings 4

2.2.1 The potential advantages of a screen on the steering wheel 4
2.2.2 Autonomous Vehicles 10
2.2.3 Human Factors For Driver-Vehicle Interfaces 12
2.2.4 UX guidelines of Android for cars 16

2.3 Discussion 20
2.4 Conclusion 21

3 Benchmarking 22
3.1 Ethical 23
3.2 Process 23

3.2.1 Steering wheel screen product or concept in the market 23
3.2.2 Functions of steering wheels in the market 24
3.2.3 Functions of the vehicle center display in the market 25

3.3 Result and conclusion 26

4 User Study 28
4.1 Define target group 28
4.2 Ethics 29
4.3 User study 1 30

4.3.1 Method and purpose 30
4.3.2 Process 30
4.3.3 Results 31

4.4 User study 2 32
4.4.1 Purpose 32
4.4.2 Method 33
4.4.3 Process 33
4.4.4 Result 34

5 Use scenario 35
5.1 Purpose 35
5.2 Scenarios 37
5.3 Analysis 38

6 Design goals 39

7 Physical Concept Generation 40

8 Concept I 48



8.1 Wireframe 50
8.2 Low fidelity prototype 1 52

8.2.1 Home page 53
8.2.2 Gmail 53
8.2.3 Youtube 56

8.3 Low fidelity prototype 2 59
8.4 High fidelity prototype 60

9 Concept II 66
9.1 Wireframe 66
9.2 Low fidelity prototype 1 68

9.2.1 Touchpad screen 68
9.2.2 Center screen 69

9.3 Low fidelity prototype 2 76
9.3.1 Gmail 78
9.3.2 Youtube 79

9.4 High fidelity prototype 80
9.5 Connection of interfaces 85

10 User Test 87
10.1 Method and purpose 87
10.2 Process 88
10.3 Results 88

11 Discussion 91

12 Conclusion 95

13 Reference 97

Appendix A: Definitions for Terms Related to Driving Automation Systems 100

Appendix B: Questions in User study 1 103

Appendix C: Participants’ answers of the interview in User study 1 105

Appendix D:  User requirements statistics 113

Appendix E: Precess of User study 2 117

Apendix F: Results of User study 2 121

Apendix G: User test questions 124



1 Introduction

1.1 Background
With the development of AI technology and deep learning algorithms, as well as the arrival of
the Internet of Things era, autonomous vehicles are the general trend. People are looking
forward to the arrival of fully self-driving cars. Self-driving cars from Level 1 to Level 3 that
are currently ready to be put into the market, the proportion of time that the system controls
the car gradually increases. As self-driving cars are likely to significantly improve road safety,
countries want to be at the forefront of this development (Vellinga, N. E., 2017). In the future,
higher-level autonomous driving technology can further reduce the distraction of perception
factors, freeing people's hands to do other things, such as checking emails, watching videos,
and reading. This will transform our experience of commuting and long-distance travel, keep
people away from high-risk work environments, and allow industries to grow and collaborate
to a higher degree.In the future, our reliance on and relationship to cars will be redefined -
reducing carbon emissions and paving the way for a more sustainable way of life. The
development of autonomous driving technology will bring different benefits and challenges to
the existing transportation system.

Compared to the advantages mentioned above, there are problems with making rules,
re-examining highway codes, public opinion, improving the infrastructure of streets, towns
and cities, and so on. Determining ultimate responsibility for road accidents is a bigger issue.
As an increasing number of road tests of autonomous driving technology are implemented,
countries have successively introduced corresponding laws to adapt to the changes brought
about by this new technology as soon as possible. On 6 January 2016, EU member states
signed a declaration on strengthening cooperation in the field of connected and autonomous
driving. Through this declaration, member states recognize the importance of a coordinated
approach to facilitate the cross-border use of connected and autonomous vehicles One of
the objectives of the Member States is (…) to work towards a unified European framework
for the deployment of interoperable connected and autonomous driving (…). This will
promote legal consistency, recognizing that the legal framework is flexible enough to
facilitate the introduction and cross-border use of autonomous and connected vehicles
(Vellinga, 2017). It can be seen from the speed of response in various regions that
autonomous driving technology has been widely concerned and researched. Many car
companies and related companies have begun to study new energy vehicles and
autonomous driving technologies and related fields.

Autoliv, the world's largest supplier of automotive safety equipment, has also conducted
corresponding research. Because autonomous driving technology provides drivers with time
and space to perform secondary tasks, there is a possibility of a screen on the steering
wheel of the car. In the past years, steering wheels in passenger vehicles include physical
buttons to trigger some functions while driving. However, the transition from manual to
automated driving is creating lots of assisting functions and features (e.g., speed limiter,
adaptive cruise control, lane-keeping assist systems, etc.), which increase the complexity of
the driving activity. Consequently, there are possibilities of replacing physical buttons for
touch screen displays. Touch screen steering wheel controls enable the possibility of

1



creating a flexible user interface, where functions and buttons can change and adapt
according to the driving context.

SAE is the leader in connecting and educating mobility professionals to enable safe, clean,
and accessible mobility solutions. The standards set by SAE represent the best technical
content developed in a transparent, open, and collaborative process. To advance automation
technology, clarify the role of the driver, answer legal questions, and clarify definitions, the
SAE On-Road Automated Driving (ORAD) Committee and ISO TC204/WG14 through a Joint
Working Group established a working group in 2018. This collaboration brought to bear the
knowledge and expertise of global experts in driving automation technology and safety.
Several new terms and definitions have been added and multiple corrections and
clarifications have been made to address frequently misunderstood concepts and improve
the utility of the document. The concept of level 3 mentioned in the article comes from the
document proposed by the organization. As a conclusion, a Level 3 self-driving car is a car
capable of self-driving in some circumstances but needs the occupant to be able to
take-over in a short amount of time. Figure 1 shows the summary of level 3 driving
automation in terms of five elements.

Figure 1：Summary of level 3 of driving automation (Made by SAE INTERNATIONAL)

While the safety implications of this concept in the context of purely assisting systems
(where the driver is responsible for the driving activity all the time) are still explored, this
concept could have potential benefits in a car with SAE L3 of automation. In this case, while
the car is performing the driving the system could provide embedded applications (e.g.,
reading emails) that allow the performance of safer secondary tasks (compared to e.g.,
using a smartphone), since the eyesight would be closer to the road and hands would
remain on the wheel. Therefore, it is worthy to explore putting a steering wheel screen on a
Level 3 self-driving car.

2



1.2 Aim and Scope
The exploration of the steering wheel screen starts with academic research in related fields
so that the knowledge gap can be filled by the study of related technologies and guidelines.
We need to study the existing similar products, discover design regulations, and design
products that meet the market and users. In addition, we need to think about the design from
the user's point of view, user study is required to obtain requirements and build analysis
models.

The next task is to come up with a conceptual graphical user interface design alternative for
the touch screen control that could provide safer secondary tasks during SAE L3
automation. Besides, evaluate them from the user experience and cognitive ergonomics
point of view.

In addition to the conceptual graphical user interface design, we must consider the
combination of hardware and software. Because most electronic systems, whether
self-contained or embedded, have a predominant digital component consisting of a hardware
platform that executes software application programs (De Michell, G., & Gupta, R. K. 1997).
To achieve a complete interactive user experience, the design of the user interface is
inseparable from the physical implementation, that is, the specific form and implementation
of the screen need to be considered. Therefore, in addition to the conceptual graphic user
interface design, we also need to design the physical level of the steering wheel screen,
explore its specific style, display mode, interaction mode, etc.

Therefore, the aim and scope of the project is to conceive and design a steering wheel
screen that meets the design specifications and ergonomics to help users complete
secondary tasks for SAE L3 autonomous vehicles, including the industrial design of the
screen and the user interface design as a combination.

3



2 Literature Review

2.1 Aim
The research field of this project is mainly in autonomous vehicles and in-vehicle
human-computer interaction systems. Although the team members have basic knowledge
and experience in ergonomics and human-computer interaction design, we lack the
background and knowledge for autonomous vehicles and in-vehicle interaction. Therefore,
the main purpose of this literature study is to fill the knowledge gap in related fields and
understand related design specifications.

Some of the specific topics that the literature study investigated are:

● Learn about the field of autonomous vehicles, especially how Level 3 autonomous
vehicles work.

● Explore the potential benefits of having a touchscreen on the steering wheel.

● Learn the design guidelines of in-vehicle human-machine interface from the
perspective of ergonomics

● Learn and analyze UX guidelines of Android for cars

In addition, in the process of literature study, research vacancies in related fields may be
found, which may have guiding significance for the follow-up research direction.

2.2 Findings

2.2.1 The potential advantages of a screen on the steering wheel

Reduce the distractions from using mobile phones

About 1.35 million people lose their lives each year due to road traffic accidents around the
world. Between 200 and 50 million people suffered non-fatal injuries from these accidents,
many of them permanently disabled. Road traffic accidents have a significant economic
impact on victims, resulting in a loss of approximately 3% of GDP for the entire country.
Supported by data, driver distraction is the biggest cause of road traffic accidents (Hua, Q.,
Jin, L., Jiang, Y., Guo, B., & Xie, X., 2021). As the most commonly used multi-functional
electronic device, mobile phones are considered to be an important factor causing driving
distraction. Drivers need to use mobile phones to complete specific secondary tasks during
driving, such as making calls on the highway, sending text messages when waiting for red
lights, etc. According to the World Health Organization, the use of mobile phones and other
electronic products quadruples the likelihood of road accidents (Chand, A., Jayesh, S., &
Bhasi, AB, 2021), and mobile phone use (texting or talking) is associated with 26% of car
crashes are related and the frequency is increasing. Cell phone use while driving is directly
related to work obligations and overconfidence in the ability to drive while talking/texting

4



(Engelberg, J. K., Hill, L. L., Rybar, J., & Styer, T., 2015). Due to busy personal affairs or
traffic jams, many drivers use fragmented time to read emails and make phone calls in the
car. Research shows that drivers text at least some of the time (30%) and at red lights (66%)
while driving on the highway (Engelberg, et.al, 2015).

The National Highway Traffic Safety Administration (NHTSA) has categorized three types of
driving distraction: visual (eyes off the road), manual (hands off the wheel) and cognitive
(mind off of driving) in their policy statement (NHTSA, 2011). So Texting while driving (i.e.
any direct manipulation of a handheld device) is considered particularly problematic because
the behavior combines manual, cognitive and visual forms of distraction. Likewise,
multi-resource theory suggests that visual-manual auxiliary tasks (such as smartphone use)
may negatively affect primary visual-manual primary tasks (such as driving a car) because
both tasks use the same cognitive resources (Naujoks, F., Purucker, C., & Neukum, A.,
2016). Therefore, when the driver looks at the mobile phone, his vision is off the road ahead,
at least one hand is off the steering wheel to hold the mobile phone, and his thinking is
separated from the actual dynamic driving task, and instead, a lot of cognitive ability is
applied to how to reply to text messages. When encountering an emergency, the driver's
cognitive ability is limited, and it is impossible to completely separate the cognition from the
second task in a short period of time, resulting in the inability to respond in time and the
occurrence of danger.

With the advancement of communication technology, autonomous driving technology, and
entertainment devices, different levels of autonomous vehicles and manual vehicles will
share the road in future mixed traffic environments. Hence, multiple factors will increase the
cognitive burden of drivers and lead to distraction (Hua, et.al, 2021). Avoiding danger to
drivers requires avoiding the use of mobile phones while driving, but the popularity of
automated driving can provide drivers with an opportunity to complete a secondary task in
certain situations. This project is based on the L3 autonomous driving technology provided in
specific situations, the completion role of the driving task requires frequent switching
between the driver and the automatic driving system, which requires the driver to quickly get
out of the distracted state to avoid danger in a timely manner. However, using a mobile
phone to complete the second task in the process of automated driving still occupies a lot of
cognitive resources of the driver, so it is worth exploring adding more interactive functions
that can complete the second task to the car itself, and the current car has come with more
functions beyond the driving function.

In order for the driver of an automated car to safely complete the second task of driving, it is
first necessary to make the driver's field of vision as close to the road ahead as possible,
thereby reducing the degree of visual distraction. The degree of manual distraction can be
reduced by keeping the driver's hands on the steering wheel as much as possible. So
exploring the steering wheel user interface allows the driver to quickly switch between their
roles (ie, action initiator and system supervisor). This is also the starting point of this project,
providing the driver with a user interface on the steering wheel that can complete the
secondary task interaction. We need to explore its possibility to ensure the life safety of the
driver of the automated car. This is in line with the goal of UN sustainable development goal
3 to reduce future fatalities from traffic accidents through the development of vehicle safety
equipment.

5



Steering wheel related factors

Since the design objective is the steering wheel, there are many factors to consider when a
touch screen is installed on the steering wheel. We should understand the basic structure of
the steering wheel to decide where to install the screen without affecting the basic functionsl.
In addition, the ergonomics of the steering wheel is also important, which affects the
operability of the screen and the user's willingness to use it. So we decided to study the
basic structure of the steering wheel and the way drivers generally hold the steering wheel.
Through the study of the steering wheel, we intend to further understand the form of the
screen and the possibilities of how to use it.

As shown in figure 2, take the steering wheel of Porsche 997 II as an example, which
consists of basic structure, steering wheel corpus, steering wheel cover with horn function
and airbag (Harrer, M., & Pfeffer, P. (Eds.)., 2017). It is well known that the airbag module in
the center of the steering wheel is a mature and effective life safety protection device, and its
position cannot be changed at will, which adversely affects the driver's safety. So the airbag
is a compromising factor in the design of the steering wheel screen, which limits the size,
material and shape of the user interface. At the same time, the car horn is currently in the
form of a bulky physical button. For the driver, requiring them to hold the steering wheel at
the 9 o'clock position with the left hand and the right hand at the 3 o'clock position or lower
prevents the airbag from pushing the driver's hand back onto the driver's body in the event of
a injury (Meschtscherjakov, A., 2017). Which is also one of the factors that limit the form of
the steering wheel screen.

Figure 2：Exploded view of a steering wheel (Porsche 997 II) (Harrer & Pfeffe, 2017)

Steering wheels have been added more and more electric functions and elements. One of
the most widely used components is the multifunction switch, for example, to operate a car
computer or a navigation system/radio. Gearbox operation/manually selected automatic or
double-tight gearboxes tend to be moved to the steering wheel, etc. In addition, modular
design and customized steering wheels are a future development trend (Harrer & Pfeffe,
2017), especially after the popularization of automated driving technology, more functions

6



are added to the car and even the steering wheel. When the steering wheel is fully
customized, it can achieve stronger inclusiveness, such as accommodating the use of skilled
left-handers, or it can satisfy drivers of different types of work. This is another reason to
explore the possibilities of steering wheel user interfaces.

Hruška, M. (2018) studied the ways and actions of the current left-hand drive car drivers to
grip the steering wheel through experiments, and expressed the proportion of the number of
people holding the steering wheel and the gender ratio in the corresponding position. From
the study we can see that drivers of different genders hold the steering wheel differently.
According to the experimental conclusion, the number of males holding the steering wheel
with one hand accounts for more than 50% of the total male subjects in the experiment, 40%
of the males hold the steering wheel with both hands, and nearly 75% of the females hold
the steering wheel with both hands. In addition, the experiment obtained the position
combination of the left and right hands of all-gender drivers holding the steering wheel, of
which R3 and L9 accounted for the highest proportion, followed by R2 and L10. Through this
experiment, we understand the user's holding habits, and the most suitable position on the
steering wheel to place the interactive interface, which can help the user to achieve the
purpose function faster and more accurately.

Figure 3：Schematic depiction of the resulting grip percentage representation values in individual
positions. Note: The values in percentages are calculated separately for the left and right hands.

In addition to the experiments on the driver's holding habits, Hruška (2018) also conducted
experiments on the body posture of the driver who gripped the steering wheel with one hand.
Holding in this position for a long period of time results in significant lateroflexie with
significant muscle strain, in particular m. quadratus lumborum, m. obliquus externus
abdominis, m. obliquus internus abdominis and m. erector spinae. Long-term driving in such
a position can lead to pain in the lumbar spine and, in extreme cases, to permanent damage
to the postural system. In the study, males using this unhealthy posture to drive are more
than females. However, the design of the steering wheel interface can affect the driver's
holding posture. Encouraging drivers to use both hands to hold the steering wheel through
the design of the steering wheel screen is a direction worth exploring. It will help more
drivers stay healthy, especially for long-term driving drivers. However, the emergence of

7



autonomous driving will inevitably free the hands of drivers, but as far as the current situation
is concerned, L3 autonomous vehicle drivers still need to frequently take over driving tasks,
which highlights the importance of steering wheels beneficial to health.

Location of visual display and controls

The location of the user interface also affects the driver's sight and operation. The location of
the in-vehicle user interface has an impact on both the driver's reception and input of
information. The suitable location of the user interface makes it easy for the driver to get
information and perform actions. Because the steering wheel is located in front of the driver,
it is one of the few things inside the car that offers excellent visibility and reachability
(Meschtscherjakov, 2017). For conditional self-driving vehicles, a touch screen control on the
steering wheel can provide drivers with a safer secondary task path in self-driving mode than
looking at a cell phone or electronic device in other places.

Current research shows that the position of a visual display has a big impact on how easy it
is for drivers to get information. With the continuous development of vehicle assistance
systems, more and more information needs to be displayed to drivers. The appropriate
location of the information display can reduce the impact of viewing the information on the
driver. For manually driven cars, reducing the frequency and time of looking away from the
road and increasing the readability of information are possible ways to reduce the impact of
displays on the driver. Human Factors Design Guidance For Driver-Vehicle Interfaces from
NHTSA approves the study by Lind, H. (2007), arguing that locating the visual warning near
the primary driving activity will enhance the likelihood of it being noticed and lower the
amount of time required to glance at that information. More specifically, as shown in figure 4,
the specification proposed by NHTSA (2016) requires that critical displays for continuous
vehicle control or critical warnings related to vehicle forward-path are located within ± 15
degrees of the central line of sight but as close to the central line of sight as practicable. For
conditional autonomous vehicles, we expect the driver to perform a secondary task while
being able to have a certain degree of perception about the driving situation and can
respond promptly when the car requests the driver to do the driving task. Although there is
no specific research at present, referring to the standard of traditional vehicles, the driver's
eyesight when performing the secondary task should be as close to the road conditions as
possible. In addition, taking the Mercedes-Benz S-class vehicle as an example, as shown in
figure 5, it will display a red light on the steering wheel when it requires the driver to take
over the driving task. If the driver's sight is right on the steering wheel when performing the
secondary task, the driver can detect the red light signal on the steering wheel and take over
the driving task promptly.

8



Figure 4: Examples of visual display locations (Made by Human Factors Design Guidance For
Driver-Vehicle Interfaces from NHTSA)

Figure 5. The button on the steering wheel will display a red light when Drive Pilot requires the driver
to take over the driving task

2.2.2 Autonomous Vehicles

This project is based on Level 3 autonomous vehicles. In order to fully understand the
autonomous driving system and its classification principles and driving methods, we studied
and analyzed the document written by SAE International: Taxonomy and Definitions for
Terms Related to Driving Automation Systems for On-Road Motor Vehicles. Below is our
summary of definitions that may be used, and analysis of Level 3 autonomous driving
systems. The detailed explanation of the definition can be found in Appendix A.

9



Definition
AUTOMATED DRIVING SYSTEM (ADS)
DYNAMIC DRIVING TASK (DDT)
[DYNAMIC DRIVING TASK (DDT)] FALLBACK
OBJECT AND EVENT DETECTION AND RESPONSE (OEDR)
OPERATIONAL DESIGN DOMAIN (ODD)
REQUEST TO INTERVENE
[HUMAN] DRIVER
[DDT] FALLBACK-READY USER

Level 3 - Conditional Driving Automation
The sustained and ODD-specific performance by an ADS of the entire DDT under
routine/normal operation with the expectation that the DDT fallback-ready user is receptive
to ADS-issued requests to intervene, as well as to DDT performance-relevant system
failures in other vehicle systems, and will respond appropriately.

Limitations from Level 3 autonomous vehicles

L3 autonomous driving technology does not fully provide convenience and safety for human
life. In an L3 self-driving car, the driving environment does not need to be monitored all the
time, but intervention requests need to be responded appropriately. The self-driving system
needs to provide "sufficiently comfortable transition times" where the dynamic driving task is
handed over from the self-driving system to manual control, in that case, the driver acts as a
backup measure to perform dynamic driving tasks. This means that the higher the level of
automation, the less the driver will focus on traffic and the system, and the less able to
regain control (Seppelt & Victor, 2016). In addition, compared to imperfect automation，
manual driving by the driver is further from perfect, with 94% of crashes attributed to key
driver-related causes such as identification errors, decision errors, and performance errors.
This seems like a classic dilemma, if we don't automate we're stuck with human
contributions to collapse, but if we choose to automate, human performance will get worse
as automation gets better.

The higher the level of autonomous driving, the more time the driver has to complete the
secondary task, so that the more cognitive resources are occupied by the secondary task, it
is more difficult for drivers to quickly switch from the secondary task to dynamic driving task.
Similarly, L3 autonomous vehicles require the driver to be the supervisor and backup of the
autonomous driving system, and also need to pay attention to dynamic driving tasks for a
certain period of time, which is different from the autonomous driving systems below L2 and
above L4. Level 3 autonomous vehicles will disguisedly increase the cognitive load on the
driver as they have to focus on so many things besides secondary tasks during autonomous
driving. This is also the reason why Seppelt & Victor (2016) promoted for changing L3
autonomous driving technology or resisting L3 autonomous driving.

10



Through a critical analysis of L3 autonomous driving technology, the current limitations of
this technology and what additional considerations need to be made during the design phase
are recognized. Although the automatic driving technology is graded according to the level of
automation, the improvement of this system level is not linear with the driver's cognitive load
and safety factor. Compared with other levels, L3 will increase the driver's cognitive load. But
compared to using mobile phones and other in-vehicle screens to complete secondary tasks
during autonomous driving, the advantage of the steering wheel screen is that the user's
hands will not be too far away from the driving task, because the steering wheel itself and its
position are not too far away from the vehicle control system and the effective field of view,
the driver can take over faster. The steering wheel screen may be a rare possibility for L3
autonomous driving technology to help users safely complete secondary tasks.

The Level 3 autonomous car produced by Mercedes-Benz

The Mercedes-Benz Drive Smart Level 3 autonomous driving feature will become available
in the S-Class luxury sedan and EQS electric luxury car in Europe by early 2022.

Germany has taken a pioneering role in this with the opening of the Road Traffic Act (StVG)
for Level 3 systems in 2017. On December 09, 2021, Mercedes-Benz became the first
automotive company in the world to meet the demanding legal requirements of UN-R157 for
a Level 3 system.

Level 3 or conditional driving automation enables the vehicle to react to its environment and
make decisions without urging the driver to take control. The autonomous driving of
Mercedes-Benz is achieved through the Drive Pilot system.Mercedes-Benz's updated Drive
Pilot system can take over the driving chores while the vehicle is traveling at the legally
permitted speed of 37 mph (60 kph). After activating Drive Pilot via a pair of haptic buttons
above the steering wheel, the system controls the driving speed and following distance while
independently performing evasive or braking maneuvers – all without physical driver
intervention.

The Drive Pilot can be activated via two buttons on the steering wheel rim. If self-driving is
available, they will turn white. If the Drive Pilot is activated, they will turn turquoise.If the
buttons in the steering wheel rim turn red, the vehicle requests the driver to retake control
within ten seconds.

The S-Class is capable of reliable conditionally automated driving in traffic jams or when
traffic density is high. The Drive Pilot controls the speed, the distance to the vehicle ahead,
and confidently steers the car within its lane. As it does so, it can also recognize unexpected
traffic situations and handle them on its own by means of braking or evasive action within the
car’s lane. However, the Drive Pilot is designed to recognize its limits of conditionally
automated driving. When the traffic environment does not conform to the conditions
predetermined by the system, the system will request the driver to intervene and regain
control of the vehicle.

Since the Drive Pilot system can perform vehicle driving tasks, the driver can have time and
energy for secondary tasks. The S-Class provides drivers with a variety of secondary tasks
including working and recreation. For example, they could communicate with colleagues

11



online via In-Car Office, answer emails, surf the Internet, or read the news. The tasks are
mainly completed through the media display.

As the Mercedes-Benz's S-Class car has been put on the market, its autonomous driving
scenarios and driving mode switching methods can provide a reference for our project. Since
this project is to design a steering wheel screen for L3 self-driving cars, it does not specify
which car it is designed for, so the existing L3 self-driving cars are considered to be the most
reasonable reference targets. Some parts of this project that do not need to be redesigned
will refer to the car's design and standards.

2.2.3 Human Factors For Driver-Vehicle Interfaces
A key element of vehicle automation is the driver-vehicle interface (DVI). DVI is the vehicle
display that provides information to the driver and allows the driver to control the entire
vehicle and various vehicle components and subsystems status control device. The safe and
efficient operation of any motor vehicle requires that the DVI be designed in a way that is
consistent with the driver's constraints, capabilities, and expectations. This document is
intended to help DVI developers of Level 2 (L2) and Level 3 (L3) autonomous vehicles
achieve these results from a human factors perspective.

The reason we chose to study this document was that it provided design guidance on the
aspects of automotive ergonomics, which complemented our knowledge. At the same time, it
can pave the way for further learning UX and UI design guidelines. Only by understanding
the background For reasons from the perspective of human factors, we can better learn and
apply design specifications.

Below is the information in this document that we think will be helpful to our project, and our
thoughts on how that information relates to our project.

General Design Guidance for Level 3 Automation

This part of the document highlights Designing Messages for Driver Comprehension. For L2
and L3 autonomous vehicles, developing and presenting messages that support accurate
and timely comprehension by the driver are considered to be the design goals. The
document proposes a design guideline for the comprehension process of drivers. Figure 6
shows what the documentation suggests to be considered at various stages of the driver's
comprehension process.

12



Figure 6：  Questions should be considered when designing messages from Campbell et al.(2018)

This document then discusses driver needs related to message complexity and identifies the
characteristics of visual, auditory, and haptic messages that affect message complexity.
Message complexity refers to the number and types of basic information elements contained
in a message, as well as the relationship between these elements. Overly complex
messages may not be properly perceived, understood or acted upon by the driver. In order
to design the presentation of information as simply as possible, while keeping the message
accurate, the document makes the following recommendations:

A. Visual Messages consist of simple icons and fonts with only the necessary detail
included. In text displays, the number of lines of text per-message is minimized.

B. Auditory Messages are simple when an immediate response is required. This could
be single or grouped frequencies presented simultaneously; such as a simple tone
that consists of a square wave.

C. Haptic Messages are simple and perceptible. Research relevant to the topic of haptic
message complexity is limited.

Based on current industry trends, it is expected that many messages in L2 or L3 automation
systems will be multisensory (presenting auditory or tactile messages accompanied by visual
messages). This topic suggests matching the format of messages to the driver's tasks,
needs, and expectations in order to improve driver comprehension and performance, and
provides guidelines for supporting design. Visual information, for example, is best for
presenting more complex information that is not safety-critical and does not require
immediate action. Audible information quickly grabs the driver's attention and can be used to
provide high-priority alerts and warnings. This means that the display on the steering wheel
is more suitable for secondary tasks unrelated to the driving task. It is important to avoid the
auditory information generated by the secondary task interfering with the driver's recognition
and understanding of important driving-related alert information.

Visual Interfaces

The visual modality is of primary importance in the driving task, and can use various sensory
dimensions, such as color, brightness, and contrast, as well as stimulus dimensions, such as

13



location, size, shape, and periodicity (e.g., flickering). Semantic content that benefits from
persistence are better suited to be displayed through visual form. In addition, characteristics
such as color, size, spacing, and temporal characteristics (e.g., flashing or noticeable
motion) can be used to maximize the prominence, legibility, and understandability of warning
messages.

The document argues for the need to place the visual interface in a location that facilitates
rapid extraction of information while minimizing line of sight off the road and negative impact
on drivability, and gives potential visual display locations as shown in Figure 7. We believe
that some of the visual displays in these examples might have a good linkage with the
display on the steering wheel we are going to design.

Figure 7：Examples of potential visual display locations
Display Locations in Image:

A. Head-Up Display
B. High Head-Down Display

C. Head-Down Display/Instrument Panel
D. Center Console

The document also emphasizes that when designing DVI, minimize glare, both on and from,
visual displays. Glare on visual displays can originate from a variety of sources in the driving
environment and can make visual displays difficult to read. In addition, the light emitted by
the display can be harsh at night, causing discomfort or, in some cases, reducing the
visibility of the external driving environment. The document also provides some strategies for
addressing such issues. For example, mitigating glare on the display in daytime driving by
providing sufficient display brightness and using high-contrast display technology to ensure
sufficient contrast. Mitigating glare that emanates from the display while driving in darkness
by displaying content with a dark background to minimize the amount of light emitted by the
display.

14



In addition, the document emphasizes the application of color. Color is a characteristic of a
visual display that can be used to convey the meaning or urgency of an alert signal.
Compared with words and symbols, color has certain advantages in the immediacy of
recognition. Here are some common relationships between color and information categories.
Red is usually associated with danger or critical situations; white is usually associated with
caution; green is usually associated with normal operation. One of our initial thoughts on this
part is that when the interactive system switches between autonomous and non-autonomous
driving modes, it may use a combination of color and other visual displays to achieve driver
attention.

The Temporal Characteristics of Visual Displays are also mentioned in this document, such
as flashing, blinking or apparent motion, to command visual attention. Some related design
guidelines are also mentioned in the document, for example, Flashing is used for important,
suddenly-occurring, situations (optimal rate is 3-4 times/s); Multiple flash mode is used for
more urgent situations (this mode uses rapid pulses of flash for each flash cycle).

Driver Inputs

As autonomous driving is an emerging topic, there is limited information on driver input
methods for autonomous vehicles. Therefore, the document has written this section based
on basic human factors information and current interface standards. Poorly designed
controls can adversely affect or impair the operation of primary driving controls. It is
important for designers to carefully consider the placement and operation of various types of
controls (Pomerleau et al, 1999; Stevens et al., 2005). Controls that are easy to understand
and operate can minimize the distraction during the transition from manual to autonomous
driving. Here are some examples of relevant design guidelines given by the document.
Controls should provide timely and clear feedback (visual, tactile or auditory) (AAM, 2006;
Bhise, 2011). Have identifiable labels (symbols or text) that are visible and located close to
the control (Bhise, 2011). The location of the controls should not adversely affect the control
of the primary driving system controls. Usage of controls must not hinder the driver's choice
to keep at least one hand on the steering wheel at all times.

Control Placement is a point in this topic that deserves a separate discussion. Designers
should ensure that control placement and operation do not interfere with driving tasks or use
other driving controls. Here are some examples of accompanying design guidelines. The
placement of controls should be easy to reach and find, and controls should be in a visible
area or be able to be found blindly (Bhise, 2011).

2.2.4 UX guidelines of Android for cars
The car screens involved in this project are all developed based on the Android system. In
order to design a user interface that meets the specifications, we need to understand and
learn the design specifications of UI and UX for the DVI of the car. Google Design for Driving
is a design hub that provides design guidelines for developers working with the two Android
for Cars systems:

15



● Android Auto: Phone-based infotainment system that’s projected onto the screens of
compatible cars

● Android Automotive OS (AAOS): Infotainment platform that car makers can
customize, build into their vehicles, and (if they're GAS partners) integrate with
Google Automotive Services (GAS)

Design for Driving foundations

At present, there is a lack of interaction design specifications for autonomous vehicles in the
market, thus it should be noted that the design principles mentioned in this part are mainly
for manual driving cars. We learned the conventional driving interaction design specifications
and extended them to apply to the user interface design of autonomous driving reasonably.
At the same time, since the L3 autonomous driving technology requires the role of users to
switch driver and the fall-back ready user, the design with reference to the manual driving
interaction design specification can make the user switch the driving role faster and safer.

Interaction principles

The interaction between the driver and the screen must be simple, non-distracting and easily
interrupted so that the driver's attention can quickly return to the road.

Keep information current and glanceable: To ensure that the information and status
presented in the interface are clear at a glance, the driver needs to quickly understand the
task or real-time updated system status by a glance at the screen. They should be able to
finish reading in 2 seconds and turn their attention to the road. Also ensure that the system
response time after user input does not exceed 0.25 seconds. If the content takes longer
than 2 seconds to load, a spinner or similar UI change should indicate that the device is
responding.

Encourage hands-on driving: Achieving safer driving requires keeping the driver's hands on
the steering wheel as much as possible. For example, avoid designing interactions that
require two hands, and design one-hand gesture interactions when necessary to ensure that
the driver has one hand to control the steering wheel and respond in time in an emergency.
A hands-free speech interface and simple voice interactions are recommended with minimal
visual and manual requirements for the driver to minimize driver distraction and help the
driver focus on the road.

Prioritize driving tasks and avoid pulling the driver’s attention away from the road for
non-essential reasons. The most important tasks a driver performs are those related to
driving – everything else must be secondary. Although we need to design the steering wheel
user interface to achieve the completion of the secondary task in the autonomous driving
process, we need to take into account the priority of the driving task, and the secondary task
needs to make compromises for the driving task at all times. And considering the position of
the touch screen on the steering wheel, it is close to the effective range of the driver's driving
field of vision. Therefore, we need to combine ergonomics and cognitive theory to study

16



carefully about the degree of how the screen occupies the driver's cognitive load in different
driving stages.

Visual principles

Make content easy to read: Content designed for car screens must be clear and easy to
read, with a consistent UI and large touch targets that drivers can identify in all viewing
conditions (day, night, bright light, etc.).

Text legibility in a driving environment can be affected by many factors, such as lighting, time
of day, font scale (thin, medium, bold) and contrast. Highly legible text helps drivers reduce
browsing time and decision-making time, thereby reducing cognitive and visual distractions.
First of all, the key information and long paragraphs should avoid using bold font, the
minimum title type characters should be 32dp, the auxiliary text size should be at least 24dp,
and the number of Roman characters per line should not exceed 120, otherwise it will
seriously increase the visual burden of the driver. In order to ensure the readability of the
text, the contrast ratios for text, icons & background should be guaranteed to be 4.5:1,
besides, the night mode is necessary and the above contrast ratio needs to be changed
appropriately. Generally, the contrast ratio is reduced in the night mode to avoid glare. It is
worth mentioning that the black background is recommended as it is suitable for both day
and night, because it is consistent with most car interior colors, and does not occupy visual
resources to cause visual disturbances during driving, at the same time, can clearly display
text and icons.

Make targets easy to touch: The smallest touch targets should be 76dp*76dp and the
minimum distance between two touchable targets should be 23dp. This prevents drivers
from being easily distracted or making mistakes when trying to touch objects that are too
small or too close to the screen.

Keep UI elements consistent: To help drivers quickly understand their screen options, the
user interface must be clear and consistent. First of all, using a consistent icon in the user
interface, the color and interaction characteristics of the same category of information can
reduce the time and cognitive effort required by the driver and makes decision-making
easier. Color is a powerful cue for reinforcing memory and recognition, such as when dialing
on the phone interface, green means initiating a call, red means hanging up, and the
corresponding green and red colors should be used in the notification window to ensure the
color consistency of the same interactive content. Users can quickly judge and react by
color.

Google Design for Driving presents basic in-car interaction principles and visual design
principles in understandable language, each of which needs to be kept in mind in our project
advancement, which is related to the safety of the people in the car. As at-a-glance
information and simple interactions are required, while self-driving cars can allow drivers to
be distracted for a while, using design to keep distraction to a minimum is what L3
self-driving technology hopes for.

17



Android Auto

As the user interface of the mobile phone is connected to the vehicle, the user can control
the app in the mobile phone by operating the car screen. Google Design for Driving gives
strict design specifications in the System Design section, including color, layout, dynamic
effects, various sizes and fonts. First of all, in terms of color, the contrast in the two modes of
day and night proposed in the Visual principles above is emphasized again, and specific
values are given for reference. In addition, the consistency of color usage is also mentioned.
The Layout section introduces the layout, pedding space and specific values for different
screen sizes, and the Sizing section specifies different button and icon sizes and bleeding
values.

Different page logic relationships should use different dynamic switching effects. The
specification divides different motion effects into

● Switching between apps
● Switching between peer views
● Extending an existing action
● Minimizing and expanding an action
● Disrupting an action

Each interactive operation needs to be supplemented with specific dynamic effects to
express the logical relationship of the pages, so that users can understand their operation
feedback more clearly. For example, in music playback software, the shared axis motion
pattern when switching from song to song in a media app reinforces that both songs are in
the same playlist.

Different user input methods are also added to the design specification. The design of
various components of the UI interface of different types of apps (media app, messaging
app, navigation app, etc.) has also been standardized. For example, the UI principle includes
how to design a standard-compliant music playback soft armor upper navigation column.

Automotive OS

Automotive OS is the advanced planted native user interfraction system, which was built into
the car itself that can be used without additional connectivity. Its design specification is
similar to that of Android Auto. However, as it is not restricted by the Android phone system,
a lot of detailed content has been added to its design specification, such as the design
principles of the dial keyboard and scrollbar.

Conclusion

Google Design for Driving places special emphasis on interaction principles and visual
principles, and is a good resource library for learning and evaluating automotive UI & UX.
The touch screen user interface on the steering wheel has many possibilities, its shape, size

18



and interaction method are very uncertain at the current stage. However,  it cannot be
designed without any design specifications, therefore the principles of interaction & vision
still need to be considered. Most of the design norms can be used for reference and
application, these design norms may need to be further developed and thought deeply by
ourselves to practice and evaluate in our design results. While some design norms are not
applicable, we need to study and evaluate more deeply then select what we need after
understanding the guidelines.

2.3 Discussion

In the above part of the article, numerous literatures helped us understand the possible
benefits of designing a touch screen on the steering wheel, and also provided us with design
guidance from the aspects of domain background knowledge, ergonomics and interaction
design. But they all have their own limitations due to the lack of solid experiments and data.

For the steering wheel screen, although we believe that it has potential advantages in
reducing driver distraction caused by mobile phone use, helping drivers maintain a better
holding gesture and providing a better location for information and control, but there is no
relevant experiment to prove that Level 3 autonomous vehicles can achieve such a purpose
with a steering wheel screen. We can't verify the reality of these potential advantages until
we have a Level 3 self-driving car and a matching steering wheel screen. But these potential
advantages can serve as the direction for our in-depth study in the follow-up research. Our
design goal is to design an interactive system that provides secondary tasks for level 3
autonomous vehicles. After our design is complete, we can use this system to design
experiments to test the potential benefits of previous assumptions. In addition to the
advantages, there may be potential disadvantages of using the steering wheel screen, such
as whether the position of the steering wheel screen is suitable for long-term viewing, and
whether the steering wheel screen and other in-vehicle displays will interfere with each other
and affect the driver's judgment? These questions are also the directions that can be
explored using the results of the project.

The background study on Autonomous Vehicles mainly helps us understand how a Level 3
autonomous vehicle works and how it differs from manual driving. We understand that ADSs
of level 3 can perform DDT only after meeting a certain ODD, but the document does not
propose what specific ODD is. This is because different ADSs have different ODD limits, and
different Level 3 autonomous vehicles may perform DDT in different environments and for
different durations. In the literature review section we cannot clearly describe the use
environment of a Level 3 autonomous vehicle, so this is of limited help in defining the use
scenarios of our product. In the subsequent definition of the use scenarios, we can only
select some common situations as the use scenarios of the product.

As for the relevant design guidelines, although they provide us with very useful design
guidelines and specifications, most of them are based on ordinary manually driven cars.
While the document from NHTSA is designed for Level 2 and Level 3 autonomous vehicles,
as is often described in the document, due to the lack of relevant experiments, more
guidance is based on experiments with manually driven vehicles. The design guidance
provided by Google also faces the same limitations, because Google provides design

19



guidance for the interaction system of ordinary manually-driven vehicles, so the
specifications in it, such as character size and other visual principles, are not for level 3
autonomous vehicles. Considering that our design is more specific to the state of the user
when not performing DDT, design guidelines to reduce user distraction while driving will not
needed to be followed or only partially followed. There is no clear judgment standard to help
us decide which specifications need to be followed, so all the design guidance must be
considered in the design process, and judgments and decisions should be made according
to the actual situation.

As a conclusion, the biggest limitation of the literature in this field is the lack of recent
experiments to provide analysis and design guidance for the interactive systems and
steering wheel screens of autonomous vehicles. Therefore, our project is of great
importance in this field. After the design results are obtained in this design project, the
follow-up research can be carried out using the design results.

2.4 Conclusion
The use of mobile phones during driving is the main factor causing traffic accidents. This
behavior combines manual, cognitive and visual forms of distraction and consumes a lot of
cognitive resources of drivers. For L3 cars and below, more frequent distractions are more
dangerous, but it's hard for people to avoid making calls or doing other secondary tasks
while driving. Level 3 self-driving technology allows the car to drive by itself under certain
conditions, giving the driver an indeterminate period of time to perform secondary tasks.
According to the definition of SAE International, the driver as a fallback-ready user needs to
take over at any time, and failure to take over in time will lead to an accident or the
termination of the journey. Therefore, the closer the driver's visual center is to the road
ahead when completing secondary tasks, the safer it will be. This is a limitation to how users
accomplish secondary tasks, and one of the technical drawbacks of L3 self-driving cars.
Level 3 autonomous vehicles require a secondary task interaction system that can quickly
disengage the driver from task content or keep the drivers’ status close to manual driving.

The original intention of the steering wheel screen is that for the driver, the position of the
steering wheel has the best visibility and reachability. If an interactive screen is provided in
this position instead of the mobile phone for the driver to complete the secondary tasks
during automatic driving, it can reduce the impact of the distraction of using mobiles, keep
the driver's attention and vision as close to the road ahead as possible, and keep hands
close to the steering wheel. When the system asks the driver to take over the car again, they
can go back to manual driving in time. Considering the structure of steering wheels and how
to hold them, our design should encourage or guide the driver to use both hands to hold the
steering wheel on L9 R3, which contributes to airbag deployment and physical fitness.

In the design stage, we need to define the secondary task according to the user's
requirements. It is necessary to consider which can be solved by the steering wheel user
interface, and which design specifications need to be referred to to design the user interface.
This is one of the difficulties of this project. Because the project involves a wide range of
fields and there is no corresponding reference basis, it has considerable freedom and
possibilities.

20



21



3 Benchmarking
Anand, G., & Kodali, R. (2008) summarize benchmarking as a continuous analysis of
strategies, functions, processes, products or services, performances, etc. compared within
or between best-in-class organizations by obtaining information through appropriate data
collection method, with the intention of assessing an organization's current standards and
thereby carry out self-improvement by implementing changes to scale or exceed those
standards. Due to the possibilities of the project and the uncertainty of the future, we need to
learn from the relevant in-vehicle interaction interface and functions of the vehicle. Since
users cannot currently experience L3 autonomous vehicles, we hardly have the possibility to
get the real experience of the completion of related secondary tasks from the user's point of
view. In the absence of corresponding real user data, studying the relevant behaviors of
users and the products that have appeared in the market or product concepts can help us
gain insight into the development direction of related technologies in the current market.
Therefore, in this project, we need to benchmark as many related vehicle interfaces and
functions as possible. For the general direction, the content we study in benchmarking is
divided into three directions, which may affect the design goals:

1. Steering wheel screen product or concept in the market
2. Functions of multifunction steering wheels in the market
3. Functions of the vehicle center display in the market

First of all, investigating the steering wheel screen concepts that are similar to this project in
the current market is required. In order to better analyze the current enterprises' unique
understanding of the given direction and different solution concepts, the following aspects
are summarized:

● Steering wheel screen size
● Location
● Interactive mode
● Connection with other screens
● Functional group
● The meaning of the screen
● Whether/what kind of self-driving car

Due to the limited number of announced and actual products currently in the market, it is
difficult to discover the design motivation and design trend of the corresponding concepts, so
we need to find possibilities in the human-machine interface of all cars. Therefore,
benchmarking in related fields is also necessary. For example, in the design phase, it is
necessary to consider which functions of the steering wheel can be integrated into the
screen; if the steering wheel screen has a subordinate relationship with other car screens,
then what kind of functions of other car displays can be controlled by the steering wheel
screen. Therefore, the market research on the necessary functions and the second task
functions of the steering wheel is the second part of benchmarking, and the research on the
functions of the current mainstream vehicle center control screen is the third part.

22



3.1 Ethical
The ethical factors that need to be considered in benchmarking is the confidentiality of the
information, so illegal channels are not applicable to obtain the confidential information of the
company. Protection of business secrets is an ethical aspect that needs to be considered
during the benchmarking. Therefore, in the process of benchmarking, the source of
information should be considered, and information that is authentic and credible and does
not affect business secrets should be obtained from formal channels such as the official
website of the enterprise as much as possible.

3.2 Process

3.2.1 Steering wheel screen product or concept in the market
At present, there are no related steering wheel screen products that have been released in
the market, but their various functions can be studied logically and reasonably through a
wide range of related concepts released by well-known companies in the automotive field.
First, we started from different companies, inquired about the latest product releases, and
summarized the concepts of the steering wheel screen into a table. The following information
in table 1 was found from the official introduction websites of the relevant companies.

BYTON m-byte Lixiang L9 Hyundai

Scale Rectangle 7’ Rectangle unknown 2 rectangles
unknown

Position Steering wheel upper
center

Middle top edge of
steering wheel

Where the thumbs are

How to interact Touch & gesture Touch Touch
Press hard for

activating

Connection between
other screens

The steering wheel
screen and the center
armrest screen are used
as the controller of the
central large screen,
Affect the big screen by
interacting on these two

Unknown Unknown

23



screens

Function groups Main menu is
customizable.
Music, maps, air
conditioning, seats,
games, photos,
meditation mode.

Touch Bar: Quickly
operate vehicle
functions

All functions in the car
– from radio volume
to seat heating

Purpose of the
screen

Apps and functions
selection, use as a

pad

Quickly operate
basic car functions Basic car functions

Is there an ADS? L3+ L4 Unknown

Table 1: Benchmarking of steering wheel screen concepts and products in the market

3.2.2 Functions of steering wheels in the market
Considering the comprehensive and valuable information in the market, it is valuable to
summarize the steering wheel functions of various vehicles from many brands. In this
benchmarking process, the main research object is the multi-function steering wheel. The
following content in table 2 comes from the official websites of the corresponding companies.

Left function group Right function group Form

Basic driving functions
Keep distance, speed
limit
Volume adjustment
slider

Media controller
Voice control switch
Heat steering wheel

Bump sensation with
panels that move ever
so slightly so that
certain features
actually feel like a
button being pushed

Basic driving functions
Speed limit, Keep
distance
Home, direction, back

Phone call
Media controller
Customize function
Voice control switch
Home, direction, back

Button feedback
display

Phone call
Media controller
Customize function
Voice control switch

Basic driving functions
Speed limit, Keep distance

Physical buttons
Small roller

24



Left function group Right function group Form

Keep distance, speed
limit

Media controller
Voice control switch

Pressing panel
buttons

Turn-signal
headlight button
Scroll： volume (up
and down), changes
the audio track (left
and right)

Car horn, wiper, voice
control switch
Scroll： "Autopilot" (press)
Pilot speed (up and down)

Touch-sensitive
buttons and two scroll
wheels

Table 2: Benchmarking of functions that the steering wheels have in the market

3.2.3 Functions of the vehicle center display in the market
Finally, we took the Tesla Model 3 as an example to study the functions of the in-vehicle
central display (figure 8). It is a unique industry benchmark in the current market as the
screen in Tesla has the most abundant functions and the best visual effects. Tesla Model 3
center screen includes basic vehicle function controllers that control the vehicle itself through
user-friendly visualization operations, such as the air-conditioning wind direction. Other
functions like Vehicle Assistance, Navigation, Music, Tricky Entertainment, Music, and Send
Messages, are also provided by the center display.

The implanted system’s interface first displays the content related to safe driving by default,
and other functional content is displayed in the form of modal windows and drawers. The
level of the operation interface is reduced as much as possible, and the content is put away
after use to significantly improve the use efficiency. The most important and most frequently
used function buttons are as close as possible to the driver's side of the display. In the
bottom function bar, the one closest to the driver's side is the Model 3 model icon. Clicking it
will call up the vehicle's detailed settings page. Because the capacity of the bottom function
bar is not enough, Tesla specially designed an application drawer and placed it in the
position of the third icon. Drivers can click it to use applications in the drawer, such as
browser, contact list, energy consumption curve, and so on.

25



Figure 8: Interface of Tesla Model 3 center screen

3.3 Result and conclusion
Through benchmarking in the current market, it can be seen that the steering wheel screen
appears in the form of concepts, and the number of related concepts is still small. First of all,
some companies in the industry tend to use the screen instead of the traditional physical
buttons on the steering wheel, so as to achieve the purpose of simplicity and full
customization to allow users to define their own steering wheel. This type of screen is mainly
responsible for quick vehicle setup on a smaller scale.

In Hyundai's concept, there are two rectangle screens on the spokes where the thumbs are
when a driver is holding the steering wheel with both hands. The screen on the left shows
the basic function settings of the vehicle. The right screen acts as a controller for the
selected function on the left. The concept first guarantees the position of the user's hands on
the steering wheel when performing the corresponding operation, making the transition
process for the user to take over the driving task shorter and the deployment of the airbag
safer. At the same time, the screen is located at the position of the index fingers of both
hands (Meschtscherjakov, 2017). In order to prevent accidental touches, the function
triggering method of the screen is also different from ordinary clicks. It needs to be pressed
hard within a short period of time to trigger the corresponding function. Although this may
reduce the operating efficiency of the driver, from the perspective of user-centered design
thinking, preventing misoperations also greatly reduces the risk of accident, so similar
triggering methods can be considered . Hyundai's concept is not specifically designed for
self-driving cars, because more attention is paid to the possibility of using a touch screen to
replace the previous physical buttons on the steering wheel.

Lixiang L9 is also a recently released car product, as for the steering wheel screen, the
physical buttons are replaced by the touch bar above the airbag cover. In this touchbar, the

26



basic control functions of the vehicle are provided. At the same time, since the vehicle
adopts the form of HUD to replace the traditional instrument panel, the steering wheel
screen also undertakes the display function of some vehicle information. On the whole, the
steering wheel screen does not help users complete the second task, but it serves as a
replacement for some physical buttons and instrument panels. However, due to the small
size of the screen, it is not conducive to the display of information, so it appears on the
steering wheel as a functional touchbar.

BYTON gave a different answer to the market. A larger screen that is not rotated with the
steering wheel is placed in the upper center of the steering wheel area. First, the screen is
used more like a tablet, with plenty of secondary task apps. The steering wheel screen also
has a frequent connection with other screens in the car. For example, during automatic
driving, the driver can project the video selected on the steering wheel screen to the center
screen and watch it together with the occupants in the car. Although the screen is fully
functional, BYTON still retains the relevant physical buttons arranged on both sides of the
screen. During automatic driving, the driver’s hands can leave the steering wheel to
complete a variety of secondary tasks by operating the steering wheel screen and the center
screen. Compared with the other two concepts, BYTON gives a more complete and effective
answer to the user's secondary task requirements during autonomous driving, giving the
driver a lot of freedom to use both hands when the vehicle is driving automatically. However,
it prevents the user from quickly taking over the driving task and increases the possibility of
injury to the user in the event of an emergency.

The three steering wheel screen concepts currently released in the market propose three
distinct solutions. The current market is not prepared enough to adapt to autonomous
vehicles. It will be difficult to find a balance between encouraging users to perform rich
secondary tasks and safer autonomous driving. Safer autonomous driving encourages the
driver to keep both hands on the steering wheel during autonomous driving, which has
obvious benefits both in terms of the speed at which it takes over the driving task and the
secondary injury of airbag deployment (Meschtscherjakov, 2017).

The steering wheel functions configured in most automotive products have obvious design
commonalities. The function groups of the latest steering wheel products are mainly divided
into two types. The first function group is responsible for setting the basic functions of vehicle
driving, including distance maintenance, automatic cruise switch, light switch, and so on.
This part of the function group is set on the left steering wheel spoke of the thumb position
held by the left hand in 4 of the 5 brands of vehicles investigated, and most of them appear
in the form of physical buttons. Another type of function group is responsible for the
secondary tasks, which generally include answering calls, voice control, volume adjustment,
and other media control functions. This type of function group appears in the form of physical
buttons on the right-hand steering wheel spoke. As a highly integrated steering wheel, Tesla
adopts a form of multi-function buttons, through the driver's pressing and rolling in different
directions, it can play different control functions. Besides, the left side function button is
responsible for the basic driving functions of the vehicle, and the right one controls the media
functions, which is similar to the market trend.

At present, cars in the market have made a clear distinction between the functions of the left
and right-hand controllers, so as to avoid confusion between the secondary task functions
and the driving functions. When considering the functions of the steering wheel screen, it is

27



also necessary to follow the general regulations of distinguishing the function groups for the
two hands. The function buttons are mainly in the form of physical buttons, which can be
operated by the driver without vision, which greatly reduces the cognitive burden of the
driver. The button feedback display, which is partially reused in the currently launched cars,
makes the entire steering wheel integrated. At the same time, in order to achieve the same
effect of reducing cognitive load and driving distraction as physical buttons. Interactive
feedback is important to this type of button, they should give users a similar experience
through realistic pressing effects and vibrations. Various interactive feedback effects need to
be considered, especially when the user uses a screen to perform corresponding operations.

4 User Study
Through literature reading and benchmarking, the corresponding design specifications and
related design concepts in the market are introduced. Then a user study is considered to be
needed, and it is one of the important design processes used to discover user needs. The
steering wheel, a product that is closely related to the user, needs a user-centered design
method to obtain the user's needs and usage problems. However, the design objective of
this project, the L3 autonomous vehicle, is at the conceptual stage in the current market, so
we cannot obtain the actual user experience and problems of users. Therefore, it is
considered to cut into the problem from the side. Although the results of user study cannot
directly reflect the needs of users, regulations can be discovered.

The user study was mainly divided into four parts (as shown in figure 9). First, reading
papers and meetings with Autoliv were used to determine the target user group for this
project. Second, interviews were used for the target group to obtain the corresponding
mobile phone and driving habits information, and discover the needs for secondary tasks.
Thirdly, through the practice of the Bodystorming method, we studied the relationship
between the user's preferences of display position and the secondary task interface. Finally,
after obtaining the results of two user studies, potential secondary task requirements were
discovered and analyzed.

Figure 9: Process of the user study

4.1 Define target group
According to literature, compared with higher and lower level autonomous vehicles, under
certain conditions and imperfect autonomous driving will significantly increase the cognitive
burden of drivers（Seppelt & Victor, 2016) because the driver needs to be the fallback-ready

28



user, who needs to pay attention to the dynamic driving task frequently. This shows that L3
autonomous vehicles have high requirements for the driver's cognitive ability and other
related abilities. In addition, Günthner, T., & Proff, H. (2021) studied the acceptance of
autonomous driving technology by users of different ages. It was found that most people
start to feel a decline in their physical ability and cognitive response-ability from the age of
50. Usually, the elderly will avoid participating in secondary tasks as much as possible for
their own driving safety due to their physical decline. Although self-driving technology is
beneficial to the elderly with declining physical ability, let them make up for the lack of
physical ability through the self-driving system. But for the market, this result means that the
usser group of older drivers, in particular, can be persuaded to use advanced driver
assistance systems by means of a suitable and user-focused value proposition. For the
elderly, higher-level and more perfect self-driving systems are needed. Besides, trust is
equally important to all user groups for the usefulness and ease of use of autonomous
driving technology, but tends to be slightly higher among younger users and is more willing
to engage in secondary tasks.

So in this project, young adults (20-35 years old) and adults (36-50 years old) with driving
skills are our primary target group, they can quickly adapt to new technology while
participating in secondary tasks with more willingness. The occupations and lifestyles of the
two target groups are different. Young adults are more willing to use mobile phones, and
their needs for secondary tasks are more youthful and trendy. Adults, as employees, have a
higher demand for information input. Therefore, the results need to be counted separately in
the following interviews.

4.2 Ethics
In order to obtain the requirements of users, it is necessary to study the daily behavior of
users, which will inevitably involve corresponding personal privacy issues, and some issues
may include the daily behavior habits of respondents. Especially for the study of driving
habits, due to the current lack of self-driving cars, it is not entirely legal for some drivers to
complete secondary tasks while driving. As legal and safe driving is admired and distracting
from driving to secondary tasks is not recommended, we need to avoid tying users' actions
to the name and identity of the individual. The interviewee's name and address information
should not be interviewed and recorded when recording the interview results, and the
interview materials should not be screen-recorded during online video interviews. Because
the protection of the personal privacy of respondents in the user research stage is one of the
ethical factors that need to be paid attention to. On the other hand, it is necessary to avoid
the description and guidance in the questions that are contrary to the laws of the country, so
as to reduce the guidance of the interview questions to the respondents on unsafe driving
habits. This helps promote safe and responsible driving. Finally, gender equality and
occupational equality also need to be considered.

29



4.3 User study 1

4.3.1 Method and purpose
In order to more effectively analyze the user requirements we need from other perspectives,
we need to use a long time to communicate with users. Only by communicating as deeply as
possible, we can discover potential needs from the behavior of users. Therefore, it is
particularly important to guide interviewees to fully recall their daily behaviors and think
deeply. Through the most suitable form of interviews, we help the interviewees to open up
their hearts and speak freely through progressive questions and more flexible time. We can
also ask follow-up questions and discuss with interviewees to further explore the needs of
users. In order to more comprehensively analyze the corresponding secondary task
requirements from other perspectives, we need to start with the mobile phone that is
currently used to complete secondary tasks to study the corresponding mobile using
behavior, including daily use and the use during driving. In addition, we also enriched the
research on user expectations in the interview, so that the interviewed drivers can first
mobilize their thinking in the form of recalling what they did this morning, then transitioned to
daily behavior, and finally imagine future scenarios. Therefore, the purpose of this interview
is mainly to obtain the user's requirements for using mobile phones, the needs of users for
using mobile phones during driving, and the user's imagined preferences and requirements
for secondary tasks during autonomous driving.

4.3.2 Process
First, we developed step-by-step interview questions, starting with the user’s travel
experience that day, asking the user to recall how they used their phone during their
commute, and why. The interviewees were interviewed on their recent long and short trips
and asked about the purpose and method of their travel, as well as how and why they used
their mobile phones on the road. Then, the questions were directed to the respondents' daily
driving habits, including the corresponding driving experience, steering wheel grip, common
functions of the vehicle center display they usually use, and the need to use mobile phones
in daily driving. Respondents were finally asked to imagine various tasks they would like to
accomplish under hypothetical autonomous driving conditions. The questions of the
interview are in Appendix B. In order to better explore user needs from the side, additional
questions will be added for different respondents and answers, which will not be included in
it.

Since the target driving user groups are divided into two types: young adults (20-35 years
old) and adults (35-50 years old), we found 5 young adult drivers for interviews and then
interviewed 6 adult drivers. The results of all interview sites will be summarized after the
interview. At the same time, we ensured the gender ratio and occupational diversity of
respondents when looking for respondents. The interviews lasted an average of 15 minutes
and were conducted in the form of an online video conference. One of the panelists is
responsible for asking questions from the list of questions, with additional questions for deep
requirements, the other panelist is responsible for recording the interview and the answers.
The interview is divided into the following steps:

30



1. Introduction of project background
2. Emphasize the protection of respondents' privacy
3. Formal interview
4. Follow-up questions
5. Feedback time and emphasize the protection of personal privacy again

Everyone's answers are included in Appendix C. In each person's answer, there is a need
for using a specific mobile app. For different secondary task requirements, we separately
counted the types of secondary tasks in different situations mentioned by the respondents of
the two target groups, and the number of times the secondary tasks were mentioned by
different respondents. Statistics on secondary tasks are included in Appendix D.

4.3.3 Results
Young adults (20-35 years old) target user groups are mainly students and young
practitioners. On the way to study and work, their main requirements for mobile phones are
navigation (including bus route apps, and map navigation), enjoying music and podcasts,
followed by news and current affairs. The use of social media software is also a raised user
requirement. This group of users has the most obvious need for music and podcasts,
although the need for navigation is the most cited. In non-work and study situations, the
respondents' needs for mobile phones are also reflected in their needs for navigation and
music/podcast. In addition, in this relaxed atmosphere, they communicate more with their
friends or use social applications to send texts. In addition, listening to audiobooks is also an
option for some young respondents.

For driving purposes, the answers given by the respondents did not show a certain pattern,
but in general, carrying items that cannot be carried on public transportation is a relatively
common reason. Most of the respondents hold the steering wheel with both hands, and
more than half of the respondents hold the steering wheel at a position higher than L9 and
R3. Their main needs for mobile phones during driving are to listen to music and radio, use
navigation, and perform control operations within the software.

Among the secondary task requirements proposed by young adult drivers in a hypothetical
autonomous driving environment, watching Youtube and playing games are two novel
secondary task requirements, while music and podcasts are still relatively basic secondary
task requirements. In addition, sending text messages, checking and replying to emails, and
using a browser to surf the Internet are also the requirements raised twice by the
respondents. Overall, Young adult respondents believe that some secondary tasks that
require high cognitive resource consumption can be completed during autonomous driving,
and current daily secondary tasks will not be replaced.

Similarly, adults have the most prominent need for mobile phone navigation during their daily
work trips, followed by music, audiobooks, and news reading. In addition, due to
interpersonal and family requirements, they frequently use their mobile phones to send and
receive text messages and make phone calls. While driving, listening to audiobooks was one

31



of the most mentioned secondary tasks, texting and listening to music were two other
important secondary tasks.

Adults drive mostly for vacation trips and family trips, followed by carrying goods. Regarding
the steering wheel holding position, the most concentrated on the L10R2. Compared with
young adults, they are more inclined to drive seriously and safely and to hold the steering
wheel more safely. They have a requirement for active input of daily information, which is
reflected in their love for reading and listening to books. Among their expected secondary
tasks, reading and podcasts remained the most prominent choices, followed by watching
videos, sending messages and emails, and listening to music and news.

In general, the two types of user groups have certain similarities in the choice of secondary
tasks in daily mobile phone use and driving. Listening to music or audiobooks is the
secondary task that occupies less cognitive resources, and was also the most mentioned
task. Whereas young people tend to appreciate music, adults tend to read. Regarding the
imagination of the two target groups for the secondary task, watching videos is the common
choice of the two target groups, and answering emails was also proposed by the two groups.
Video media and text input were the most easily associated kind of tasks among users.
Therefore, they are the secondary tasks that we decided to develop in-depth.

4.4 User study 2

4.4.1 Purpose
After obtaining the possibility of user needs for various secondary tasks from user study 1,
some questions and follow-up research directions appeared. First, the possible secondary
task requirements were estimated based on the user's mobile phone usage and the user's
speculations. So we need some way to verify which secondary tasks are actually required by
the user in an autonomous driving environment. Second, as drivers need more information
while driving, and as the car wants to provide more secondary tasks, Level 3 autonomous
vehicles will likely offer many displays.  For example,the Head-Up Display ,the High
Head-Down Display, Instrument Panel and Center Console Display may exist in the field of
vision of a driver at the same time. In this case, the screen on the steering wheel and other
displays need to be clarified in relation to secondary tasks. Display and control are
considered to be the two most important relationships between secondary tasks and
displays. Therefore, for vehicles with multiple displays, the responsibility of the screen on the
steering wheel for secondary tasks and the relationship of secondary tasks to multiple
displays needs to be defined.

In summary, the overall purpose of further user studies is to validate the results of the
previous user study and explore users' thoughts on the relationship between secondary
tasks and multiple displays.

32



4.4.2 Method
Different from other experiments that study the user experience of drivers in ordinary cars,
we hope that the environment in which the experiment is located is a level 3 autonomous
driving environment. This means that the user can see the driving environment outside the
car while sitting on the driver's seat, but at the same time, they do not perform driving tasks
for a period of time. So far, there is no widespread Level 3 autonomous vehicle on the
market, nor is there a way to simulate a Level 3 autonomous driving environment. In talking
with Autoliv's supervisor we learned that Autoliv has driving simulators that simulate both
autonomous and manual driving scenarios. This simulator is considered to be able to help us
do better research. An experiment can be designed based on the simulator. And
bodystorming is considered the most suitable user research method for this experiment.

Bodystorming is an immersive innovation process that involves role-playing and physical
involvement with props, prototypes, real items, and physical environments to explore ideas.
Its overall aim is to comprehend the interactions between people, their physical location, and
the items (e.g., tools, devices, materials) they employ in that context (Wilson, 2011).

The main implementation method of this experiment is that the simulator provides a physical
environment and an electronic driving environment, and we supplement the environment
details by descript, including time and tasks. We print pictures of all possible secondary
tasks and their different functional components, show them to the user after simulating
entering self-driving mode in the simulator, and let them choose their preferred secondary
task and perform the interaction they think makes sense. We observe and record from the
sidelines.

4.4.3 Process
According to the target group, we found 3 young adult drivers and 3 adults drivers for
physical and online experiments. In order to keep the information of the participants
confidential, we did not record and take pictures in the physical experiments and online
experiments. The average duration of each experiment is 15 minutes, and the following is a
simplified version of the experimental process. The full experiment procedure is attached in
the appendix E.

We followed the same process for both physical experiment and online experiment. First is
the introduction of the project and the experiment. After that, we described a detailed
scenario for users helping them immerse themselves in the scenario and make decisions.
Figure 10 shows how we create scenarios through Miro. They were told to imagine they are
driving a Level 3 self-driving car with many screens, including a touch screen on the steering
wheel. Then they were required to choose three secondary tasks they want to perform
during autonomous driving from the secondary tasks we provide. For each secondary
task,they were asked to place different functional components where they see fit. Take email
as an example, a reading email page, a writing email page, and a keyboard were provided.
Users may believe a bigger screen is more suitable for reading email, and the steering wheel
screen is more suitable for a keyboard. And they were asked what kind of touch screen on
the steering wheel they would like to have. At the end of the experiment, we invited
respondents to ask us questions and give feedback.

33



Figure 10: Scenario created by Miro

4.4.4 Result
Although we used onsite and online experiment, because we have the same purpose and
process for these two methods, we comprehensively analyze all the results of the two
methods. The complete record of this study is attached in Apendix F.

Combining the experimental results of the two target groups, some useful conclusions are
found. First, we wanted to see if the experimental results showed that some secondary tasks
were mentioned significantly more than others. Combining the responses of all participants,
we found that Spotify, Messenger, Youtube, and Gmail were mentioned the most. But due to
our small number of participants, this result does not indicate that some secondary tasks are
especially valued and favored in the self-driving scene. Larger and longer studies may be
required to obtain data on secondary task selection. Then for our design, the conclusions of
user study 1 are more worth considering when deciding which secondary task should be
provided.

Regarding the placement of different functional components in the application, although
participants chose different types of applications, we found some commonalities. Figure 11
shows the components that we concluded were the most frequently selected by the
participants to be placed on the steering wheel screen. Among them, the play function
buttons and keyboard are considered to be the most suitable components to be placed on
the steering wheel screen. We believe that users tend to place components with control
functions on the steering wheel screen, such as the play and pause buttons in playing video.
Components with reading or watching attribute are more placed on other larger screens,
such as video or email content. In addition, users believe that some content can be
displayed on the steering wheel screen and other screens at the same time, which will give
users more freedom. This inspired us to think about the relationship between multiple
displays in the car. Different alternatives can be proposed later in the design phase to
explore multiple possibilities.

34



Figure 11: Components that most frequently selected by the participants to be placed on the steering
wheel screen.

Additionally, no more consistent preferences were found in the user's perception of the
position of the steering wheel screen. But some concerns of users were found. For example,
users are more concerned about whether the position of the steering wheel screen provides
a comfortable operation and whether they need to keep their heads down to observe the
screen. What this inspired us is that users believe comfort and ease of use come first, and
we should focus on exploring good controls.

5 Use scenario

5.1 Purpose
After obtaining a lot of user behavior and preference information, we hope to use a more
vivid way to summarize and analyze the obtained information, and to help us better predict
the user's intention and performance when using the products we provide. Persona and use
scenarios are considered to be a good way to help us translate user needs into product
concepts. Since we have two target user groups, we want to create two persona and their
use scenarios according to their respective typical needs, in order to simulate and explore
common use scenarios of all target user groups.

Persona is a fictional character created from research to represent possible different types of
users. Creating personas will help users understand their needs, experiences, behaviors,
and goals. (Dam & Siang, 2022). Nielsen (2013) proposed a method “Ten steps to personas”
to construct persona, as shown in Figure 12. This process model contains four different main
parts: data collection and analysis, persona descriptions, scenarios for problem analysis and
idea development, acceptance from the organisation. Nielsen (2013) argues that a project
does not need to follow all steps. In this project, the sixth step “defining situations” is
considered to be the most efficient approach and we call it creating use scenarios. We want
to create some virtual characters and usage scenarios that describe the user's scenario

35



when using our product, including the user's vision and behavior, possible feedback and
performance of the product.

Figure 12: “Ten steps to personas” to construct persona by Nielsen (2013)

We believe this is critical step in the project, because the creation of the use scenario is
based on the results of the user research, which represents the patterns of user behavior
and needs found in the user research, and is the summary and presentation of the data in
the user research phase. At the same time, the use scenario is also the beginning of the

36



product concept generation. In this step, we define the use environment of the product in
more detail, the scope of the design, and the results we want to achieve.

In the user research stage, we found that media and text input are the most desirable
secondary tasks for users, and they are also the two types of secondary tasks that we
decided to develop in depth. Therefore, when creating use scenarios, these two types of
applications are integrated into the story, and the email application and video application are
highlighted.

Another reason to choose these two types of apps is that video apps are more of an
entertainment and relaxation attribute, while email apps are more of a business attribute.
Reflecting on the secondary tasks that users frequently mentioned in the user research they
wanted to perform in a Level 3 autonomous vehicle, some of them were more for
entertainment purposes, and some were more for office purposes. This gave us the hint to
create a relaxation scene and an office scene when creating the use scenario. Therefore,
reading and replying emails, and watching videos, respectively, match the two usage
scenarios that users expect from a Level 3 autonomous vehicle. According to the results of
user research, young adults use more entertainment and relaxation apps on a daily basis, so
we designed a scene that is biased towards relaxation for young adults. Adults, on the other
hand, showed preferences for self-improvement and work in user research, so we designed
an office scenario for adults.

To make the story seem more real and specific, we substitute each type of app with a
specific product that is most widely used in that type of app. For example in Young adults'
use scenario, we use Youtube instead of "video app".

5.2 Scenarios
For young adults, we created a couple's persona and a situation where they drive to IKEA to
shop.

Carl (26 years old) and Anna (25 years old) are students at Chalmers University of
Technology and live in the city centre of Gothenburg. On a weekend in April, Carl and Anna
decide to go to IKEA to buy some furniture and decorations for their home . Since IKEA is so
far away