Examining factors for low use behavior of Advanced Driving Assistance Systems

(1)

Linköping University | Department of Computer and Information Science Master’s thesis, 30 ECTS | Cognitive science Spring term 2020| LIU-IDA/KOGVET-A--20/006--SE

Examining factors for low

use behavior of Advanced

Driving Assistance Systems

Kajsa Emanuelsson

Supervisor: Erik Prytz Examiner: Arne Jönsson

External supervisor: Annika Larsson

Linköping University SE-581 83 Linköping, Sweden +46 013 28 10 00, www.liu.se

(2)

(3)

Upphovsrätt

Detta dokument hålls tillgängligt på Internet – eller dess framtida ersättare – under 25 år från publiceringsdatum under förutsättning att inga extraordinära omständigheter uppstår. Tillgång till dokumentet innebär tillstånd för var och en att läsa, ladda ner, skriva ut enstaka kopior för enskilt bruk och att använda det oförändrat för ickekommersiell forskning och för undervisning. Överföring av upphovsrätten vid en senare tidpunkt kan inte upphäva detta tillstånd. All annan användning av dokumentet kräver upphovsmannens medgivande. För att garantera äktheten, säkerheten och tillgängligheten finns lösningar av teknisk och administrativ art. Upphovsmannens ideella rätt innefattar rätt att bli nämnd som upphovsman i den omfattning som god sed kräver vid användning av dokumentet på ovan beskrivna sätt samt skydd mot att dokumentet ändras eller presenteras i sådan form eller i sådant sammanhang som är kränkande för upphovsmannens litterära eller konstnärliga anseende eller egenart. För ytterligare information om Linköping University Electronic Press se förlagets hemsida http://www.ep.liu.se/.

Copyright

The publishers will keep this document online on the Internet – or its possible replacement – for a period of 25 years starting from the date of publication barring exceptional circumstances. The online availability of the document implies permanent permission for anyone to read, to download, or to print out single copies for his/hers own use and to use it unchanged for noncommercial research and educational purpose. Subsequent transfers of copyright cannot revoke this permission. All other uses of the document are conditional upon the consent of the copyright owner. The publisher has taken technical and administrative measures to assure authenticity, security and accessibility. According to intellectual property law the author has the right to be mentioned when his/her work is accessed as described above and to be protected against infringement. For additional information about the Linköping University Electronic Press and its procedures for publication and for assurance of document integrity, please refer to its www home page: http://www.ep.liu.se/.

(4)

Abstract

Advanced Driving Assistance Systems (ADAS) has the potential to decrease the number of fatal accidents in traffic. However, in some cases, drivers with the systems in their car are resistant against using them. Exploring the underlying reasons and factors of the low-usage of ADAS was the purpose of this thesis. The thesis consists of Study I, an exploratory interview study with ten drivers who had cars with ADAS. The goal of Study I was to highlight the possible reasons behind the low usage of ADAS. The results of Study I were used to design Study II, which consisted of a survey targeted to drivers who had access to the ADAS adaptive cruise control and lane keep assist (N = 49). The results indicate that the factors or circumstances that affect usage depend on the ADAS and the user groups. Some identified underlying factors for low usage behavior of ADAS are the need to monitor the vehicle more when ADAS is activated and lack of trust in own ability when using ADAS compared to the high usage group.

Keywords: ADAS, Adaptive Cruise Control, Lane Keep Assist, low usage

Sammanfattning

Advanced Driving Assistance Systems (ADAS) har potential att förhindra antalet dödsfall i trafiken. Det förekommer att förare som har systemen i sin bil, väljer bort att använda dem. Syftet med den här uppsatsen var att undersöka underliggande orsaker och faktorer till låg användningsgrad av ADAS. Uppsatsen består av två studier. Studie I är en explorativ intervjustudie med tio förare som hade bilar med ADAS. Målet med Studie I var att ringa in de möjliga bakomliggande faktorerna för låg användningsgrad av ADAS. Resultaten från Studie I användes för att utforma en enkätstudie till Studie II som var riktad till förare som hade bilar med förarstödsystemen adaptiv farthållare och körfältsassistans (N = 49). Resultaten pekar på att de underliggande orsakerna och faktorerna beror på vilken ADAS som avses samt vilket användargrupp föraren tillhör. Några underliggande faktorer för låg användingsgruppen tycks vara känsla av att behöva övervaka fordonet samt lägre grad av tilltro till den egna förmågan än vad höganvändingsgrupper rapporterade.

(5)

Acknowledgment

This thesis is conducted at the request of Veoneer. I would like to express my gratitude for the opportunity to write a thesis connected to research and development within an industry. It has been a valuable experience. Furthermore, I would like to thank my supervisors Annika Larsson, and Erik Prytz, for your guidance and wise thoughts. Especially when the current situation in the world required a resilient mindset. The thesis would not have been possible without the participants. Thank you very much for taking the time to help me in the writing of my thesis. Particularly, the pairs in the interview study. This would never have been possible without your valuable input and opinions.

Thanks to my father for letting me, in a time of distance mode, establish a home office at his house, and for his thoughtful, though not always fruitful, attempts to make me call it a day. Finally, thank you, Martin, for your encouragement, seemingly infinite patience, and grammatical skills.

Kajsa Emanuelsson Göteborg, 2020

(6)

(7)

List of Abbreviations

Abbreviation Meaning

AAP Active Accelerator Pedal

ACC Adaptive Cruise Control

ADAS Advanced Driving Assistance Systems

AEB Automatic Emergency Brake

BLIS Blind Spot Information System

FCW Forward Collision Warning

ISA

Intelligent Speed Adaptation

LDW Lane Departure Warning

LKA Lane Keep Assist

(11)

(12)

1

1 Introduction

Road accidents are the 8th most common cause of death in the world with 1,35 million lives lost every year (World Health Organization [WHO], 2018). One of the strategies to decrease the number of accidents is to support drivers with Advanced Driving Assistance Systems (ADAS). The driving support systems aim to enhance driving performance. Either by informing the driver, by warning, or by intervening and thereby relieving the driver from some of the driving tasks (Van Driel, 2007). Examples of one common ADAS is adaptive cruise control (ACC), which assists the driver in maintaining a set speed and the distance to the vehicle ahead. ACC is a system that the driver must activate and deactivate when driving. However, some systems such as the forward collision warning (FCW) is in most cases activated in the background (Reagan & McCartt, 2016).

The included systems in a car equipped with ADAS can vary but typically it includes for example adaptive cruise control and lane-keeping assistance. One of the goals of ADAS usage is to decrease the accidents caused by human fallibility. Today, the majority of car accidents are caused by what is sometimes called human error (Kalra & Paddock, 2016). Therefore there is potential to save lives if these factors can be minimalized (Bekiaris, 2011). Namely, some common causes of fatal car accidents entail sleepiness, distraction, and intoxication. These factors respectively contribute to 2.5, 10, and 41 percent of all fatal car accidents in the United States (National Highway Traffic Safety Administration, 2011; Bureau of Transportation Statistics, 2014; U.S. Department of Transportation, 2015). According to Kalra and Paddock (2016), the performance of a car with ADAS may also be superior to a manual car due to better decision-making, better perception, and faster execution of tasks.

Even though the systems are designed to provide a safer and more comfortable way of transportation, and even though the car companies spend a considerable amount of money to develop ADAS, not all owners of cars with ADAS use the systems. Especially not the ones they have to activate themselves. Through the knowledge of why drivers neglect the possibility to use ADAS, there will be an increased potential to adapt the systems accordingly and make the drivers use the systems available to them. Therefore, it is important to know why a driver is resistant to using the systems. There has not been a lot of research regarding the underlying reasons for none or low usage of ADAS. Therefore, it is identified as a gap in research which this thesis aims to investigate.

(13)

2

1.1 Purpose

The purpose of this thesis is to examine the underlying factors or circumstances that regulates a driver’s usage and non-usage of Advanced Driving Assistance Systems. Underlying factors could include driving style, social and cultural factors. Circumstances include inside- or outside car context. Context is used to describe the situation where the event is taking place. Outside car context can include road quality, weather, or traffic density. Inside car context covers what is happening inside the vehicle, for instance how many people are inside the car, the behavior of the passengers, and the relationship between the driver and the passengers. The purpose is not only to identify the factors but to investigate which factors play a more significant role than others when a driver is deciding whether to use ADAS.

1.2 Research questions

Based on the purpose of the study, the following research questions were established:

RQ1: Which factors or circumstances are most important for the driver when deciding whether to use ADAS?

RQ2: Are there differences in the usage of ADAS between user groups?

RQ2.1: If there are differences in usage between groups, what are the characteristics of these groups?

Due to the exploratory approach of the thesis, there is not a formal hypothesis. However, it is expected that there will be some underlying factors or circumstances that are more important than others when a driver is deciding whether to use ADAS or not. The purpose of this thesis is, therefore, to find which factors are more influential than others.

1.3 Limitations and Demarcations

The participants of this study were all from and living in Sweden. Therefore, care should be taken when generalizing the results to other countries or cultures. According to Planing (2014), there are differences between how cultures perceive the potentials of ADAS. This paper only aims to investigate the Swedish perspective.

Furthermore, the average owner of a car with ADAS is likely not a representation of the average car owner. This due to that cars with ADAS are in general more expensive than cars without, and therefore it is possible that the owners of ADAS equipped cars have an economic situation that allows them to buy that a car of that category. However, this paper does not claim to make a statement regarding car owners in general and therefore this is not viewed as a major issue.

(14)

3

2 Theory

This section will present theoretical background regarding the level of support, ADAS function description, activation and deactivation of ADAS, effects of ADAS usage, driving style, learning strategies for ADAS, and acceptance.

2.1 Level of support

The driving support system can be divided into different levels of support depending on how the system is related to the driving task. The levels can be divided into:

Inform the driver

For instance, by letting the driver stay updated about the current speed limit. Assist the driver

Suggest how the driver should act in certain situations (Van Driel, 2007). This can occur for instance through a signal indicating that there is an obstacle when driving in reverse. The purpose of the signal is to avoid accidents, but the suggestion can be overridden by the driver. The information or warning that the system is trying to get through can take multiple forms. For example, through lights, sounds, tactical feedback (Brookhuis, de Waard, & Janssen, 2001) Control an aspect of driving

Lastly, a system can be controlling of the driving situation (Van Driel, 2007). In that case, the system makes the decisions and implements them. For instance, when the system adapts the speed of the vehicle according to the vehicles in front of the car.

2.2 ADAS function description

The support systems can be divided into lateral and longitudinal systems. In this section, a closer description of some of the most common ADAS is presented.

2.2.1 Lateral Systems

The lateral driving assistance addresses safety on the sides of the vehicle. For instance, to keep the vehicle within the sideline markers or to assist when the driver is moving the vehicle sideways. Examples of the lateral systems and their functionality are presented below.

Lane Departure Warning

The Lane Departure Warning (LDW) will inform the driver if their vehicle leaves the current lane. The purpose is to decrease the number of accidents where the driver unintentionally leaves the lane or leaves the lane without first using the turn signal. There are multiple methods to determine the position of the vehicle. Most commonly through a camera but it is also possible through satellite positioning or magnetic markers in the roadway (Van Driel, 2007). If the systems notice that a sideline marker is surpassed the driver will be informed.

Lane Keep Assist

The Lane Keep Assist (LKA) is based on the LDW, but if the sideline markers are surpassed the system will steer the driver back into the lane. The system requires the driver to keep the hands on the steering wheel otherwise it will deactivate. It is therefore not meant to be used as an autopilot function.

Blind Spot Monitoring

Blind Spot Monitoring assists the driver by informing if there is a vehicle in the driver’s blind spot or passing the driver in an adjacent lane. The system uses cameras to notice when a nearby vehicle causes an unsafe lane-change transfer situation. If a vehicle is detected, the driver is

(15)

4 warned either through an audio or visual signal. In Volvo’s Blind Spot Monitoring, called BLIS, the warning is indicated through a light in the side mirror (Volvo, 2020).

2.2.2 Longitudinal Systems

The longitudinal systems support the driver in traffic situations related to moving the vehicle forward. The relevant aspects are for example maintaining an adequate speed and keeping enough distance to other vehicles.

Adaptive Cruise Control

The Adaptive Cruise Control (ACC) assists the driving in adapting the speed in accordance with the vehicle ahead. The driver can set a speed that they would like to go, but also a minimum time interval to the vehicle ahead. Through radar sensors, the system scans the traffic flow. If a vehicle ahead is driving more slowly than the speed set by the driver, the speed will automatically decrease to match the vehicle ahead. If the vehicle ahead starts to drive faster or leaves the lane, the vehicle will resume the previous desired speed. The ability to adjust to how fast other vehicles are driving is what makes the system adaptive compared to a conventional Cruise Control that maintains the desired speed regardless of the surrounding environment. The ACC is according to Volvo (2020), not made for city traffic, heavy traffic situations, slippery roads, during heavy rain, or other demanding driving conditions.

Forward Collision Warning

Forward Collision Warning (FCW) warns the driver if an impending crash situation is detected. There are different methods to discover potential crash situations. For instance, through camera and radar sensors. The purpose of the system is to discover a possible risk of collision with a pedestrian, cyclist, or another vehicle.

Automatic Emergency Brake

In contrast to the FCW, the Automatic Emergency Brake (AEB) will brake the vehicle if there is an impending crash situation. Therefore, the AEB is an intervening system in comparison to FCW. According to Volvo (2020), it is not possible to deactivate the AEB in their cars. However, it is possible to deactivate the warnings, i.e., the FCW.

Pilot Assist

This type of driver assistance has different names depending on the car brand. In Tesla’s it is called Autopilot, Mercedes-Benz has named it Drive Pilot, and in a Volvo, it is called Pilot Assist. The function is a combination of the ACC and the LKA. Therefore, it maintains and adjusts speed and keeps the vehicle within the current sideline markers. However, the driver must keep the attention on the road and is suggested to keep the hands on the steering wheel. Otherwise, the system will warn the driver.

Backup camera and sensor

The systems assist the driver by communicating a warning if there are any obstacles behind the vehicle. It also provides information through a visual image of the situation behind the vehicle. The system is active when the car either has the reverse gear engaged or if the car begins to roll backward.

2.3 Activation and deactivation of ADAS

Having a specific ADAS does not automatically lead to usage. In a study by Harms, Bingen, & Steffens (2020) they asked drivers what ADAS they had in their current car. Five out of seven drivers who knew that they had a specific ADAS also reported using it. The authors reported being aware of owning a system as an essential factor for usage. They also found that not all

(16)

5 drivers were aware of the ADAS they had in their current car. Braitman, McCartt, Zuby, and Singer (2010) found that 25 % of drivers with LDW did not know that they had the system in their car. Even if the drivers are aware that they have LDW, they might still deactivate it. Braitman et al (2010) found lower usage of ADAS that requires manual activation. Some ADAS, such as the AEB, does not require activation since it is a default activated system. Therefore, awareness and knowledge of the system may not be of equal importance compared to other ADAS. This, in contrast, to e.g., the ACC where activation from the driver is required. Reagan and McCartt (2016) conducted a study where they observed Hondas brought into dealerships for service or repair. Out of the 265 vehicles they observed, 32.8 % had LDW activated. They found a difference in the likelihood of activation, depending on the model of the car. For instance, in Honda Accord it was 66 % more likely to see an active LDW than in a Honda Odysseys. Except for control of LDW, they observed the activation of FCW and found that 264 vehicles (99.6 %) had FCW activated.

It is important to understand what factors cause a driver to activate systems. Pauwelussen and Minderhoud (2008) found that drivers would deactivate the ACC to overtake a vehicle ahead. When the drivers deactivated the ACC, they would keep a smaller distance to the lead vehicle. The use of ACC depends on the surrounding environment according to Viti, Hoogendoorn, Alkim, and Bootsma (2008). In an observational study, they found that drivers primarily use ACC when driving in low-medium density traffic and when driving further distances.

2.4 Effects of ADAS usage

As with all new technology, risks must be considered. From a strict technological viewpoint, that could entail random or systematic failure (Bekiaris & Stevens, 2005). Besides, there are potential issues that can occur when the driver is interacting with the systems. The driver can for instance misunderstand to what extent the vehicle oversees the driving situation. This is referred to as an Automation Expectation Mismatch (Victor et al., 2018). The mismatch can be caused by the irony of automation (Bainbridge, 1983), namely that the better and more reliable the automation of a system is, the less attention the driver pays to the surrounding traffic and to the system itself. Therefore, the driver will be slower if they need to regain control over the vehicle (Victor et al., 2018). This can cause longer reaction times and lower driving performance. Heikoop et al. (2019) claim that the amount of skills required to drive today is higher than ever before. The demand for skills is higher today due to the introduction of new technology in cars that the driver must know how to use. This new technology could lead to stress in drivers since they are expected to perform new tasks, such as learning how to use an ACC or program a GPS (Smith et al., 2008; Planing, 2014). An issue with the vehicles equipped with ADAS is according to Planing (2014), that the purchase of such a system does not include any training, except for the training received when learning how to drive. For some drivers, decades have passed since learning how to drive which often was before they had the economic resources to buy a car with ADAS. New inventions in safety systems tend to start in more expensive cars and it could take a while before the systems end up in the car models of the average driver (Planing, 2014).

The usage of ADAS has been shown to alter driver’s behavior in traffic, both in positive and negative ways. ADAS-usage will according to Brookhuis, de Waard, and Janssen (2001) lead to multiple positive effects such as considerable enhancement of driving safety since the systems will be able to, some extent, compensate for the driver’s limitations. Automation can also lead to a smoother flow in traffic, which will accommodate more vehicles on the road at a given moment. ADAS can also improve driving performance, even when driving conditions are

(17)

6 not optimal. This is especially important for elderly and novice drivers. It can for instance help elderly drivers in determining how fast other vehicles are approaching an intersection, notice traffic signs and signals, and react quicker in a potentially dangerous traffic situation (European Union [EU], 2016). Moreover, Brookhuis, de Waard, and Janssen (2001) believe that possible negative consequences of ADAS include lower perception of risk, overestimation of the ADAS capability, and disengagement. Price, Lee, Dinparastdjadid, Toyoda, and Domeyer (2019) found, in a simulator study, that drivers who were told that they were primarily in charge of driving kept their eyes on the road more than drivers who were told that their driving was assisted by automation. The type of driving that is assisted by automation requires the full attention of the driver as the driver also oversees monitoring automation (Solís-Marcos, 2019). In another simulator study, Solís-Marcos, Galva-Carmona, and Kircher (2017) found a decrease in drivers' mental attention allocation after driving partially automated vehicles. One possible aspect to be affected by mental attention allocation is reaction time. Larsson, Kircher, and Andersson Hultgren (2014) found in a simulator study that with ACC usage there was an increased brake reaction time. This was measured on instances when a vehicle cut-in from another lane and became the ACC systems closest front-vehicle. However, more experienced ACC-user had a shorter reaction time than novel users. This could indicate that drivers with experience of ACC have a greater understanding of the functionality and limitations of the system.

How drivers change their driving routine after integrating new systems in their everyday life is referred to as Behavioral adaption (Reinmueller & Steinhauser, 2019). What the behavioral adaption will entail can sometimes be hard to predict. In a field study by Lai, Hjälmdahl, Chorlton, and Wiklund (2010) drivers had an ISA (Intelligent Speed Adaptation) installed. The ISA would inform the driver about the current speed limit. The results indicate that overriding the ISA suggestions seems to increase with experience. However, when the study ended after one year not all participants showed a stabilization in their overriding behavior. In a similar field study, Wallén Warner, and Åberg (2008) discovered results that suggest that the driver’s overriding behavior with an ISA may take three years before stabilizing. Lai et al. (2010) emphasize that the ISA is an ADAS that is activated by default. Therefore, exposure to the system is more intense than with a system the driver must activate themselves, such as the ACC. Therefore, the adaption to a non-default ADAS might take even longer than to a default ADAS since the driver would only be exposed when the system is activated.

There have been multiple studies on how ADAS usage affects the planning of drivers. For instance, Rajaonah, Anceaux, and Vienne (2006) found that drivers using ACC start the overtaking initiative earlier than the drivers that do not use ACC. This would indicate more extensive planning in ADAS users. However, there might be other not equally favorable consequences of behavioral adaption to ADAS. Hoedemaeker and Brookhuis (1998) showed in a simulator study that drivers using ACC drove faster than the non-ACC group. However, it is important to remember that these are the results of a simulator study. It might be the case that drivers adjust their driving when they knew that no lives were at stake. On the other hand, in another study, the use of active accelerator pedal (AAP) has been found to promote speed reduction (Várhelyi, Hjälmdahl, Hydén, & Draskóczy, 2004). The AAP gives the accelerator pedal a counterforce if the driver tries to override the speed limit. However, the driver can still surpass the speed limit if the accelerator pedal is pressed hard enough. The results showed a decrease in speed compared to other cars driving on the same roads. In conclusion, there are different studies, often with different methodological approaches that indicate different types of behavioral adaption with ADAS usage.

(18)

7

2.5 Driving Style

There are multiple definitions of driving style. The driving style can for instance, be affected by context, such as the driving environment. Driving style is according to Itkonen, Lehtonen, and Selpi (2020) defined as the descriptions of individual variability in driving behavior.

If a driver is more prone to taking risks will likely affect the driving style. A more risk-taking driver could for instance be more likely to overtake other cars or get impatient when surrounding cars drive slowly (Watson-Brown, Scott-Parker & Senserrick, 2019). Possibly, risk-taking is linked to the personality of the driver in general (Sagberg, Selpi, Bianchi Piccinini, and Engström, 2015). Other personality traits that could affect the driving style are for example aggression and neuroticism. Aggression in traffic can be expressed through e.g., inappropriate honking or speeding. To some extent, aggressive driving could lead to risky driving behavior. If a driver has a high level of neuroticism it might cause the driver to feel insecure about driving and become highly emotionally affected when something unforeseeable happens. For example another car cuts in very close. Einarsson, Erhardsson, Stensson, and Wendt (2019) found that calmer drivers are more likely to use ADAS.

Social and cultural aspects can also affect one's driving style. These aspects can probably in extension have an impact on personal factors. The road safety values embedded in the culture seem to have an impact as well (Sagberg et al, 2015). According to Itkonen, Lehtonen, and Selpi (2020), driving style is not only a product of the conscious choice but also a result of how skilled the driver is. Factors such as experience contribute to increasing the driver’s skill level of driving. A higher level of experience can be a product of driving often and driving in different types of environments.

The driving style can vary due to the surrounding environment, in other words, the driving context. Influencing factors can, for instance, be the weather, road conditions, and traffic density. Drivers will probably act differently depending on the status of the driving context (Orlovska et al., 2019). In the category of driving context are also the technical aspects included. Unfamiliar technical settings might affect the style of driving. Furthermore, if the driver is not comfortable with the technical settings of the car it is unlikely that the driver will make use of all the technical functionalities (Einarsson et al., 2019).

2.6 Learning Strategies for ADAS

When promoting ADAS usage it is important to remember the relevance of how the system is presented. Poor ADAS training might cause knowledge gaps regarding the ADAS functionalities and hence the awareness of the ADAS present in the car. According to the production paradox, the user will not use a new feature until they want to get something done, which the feature in question can help them with (Carrol & Rosson, 1987). Even if it is the case that they want to use a tool it might not be enough. Einarsson et al., (2019) found that novice ADAS users were not able to activate ADAS without assistance. The participants found it difficult to figure out what the car was trying to communicate.

To increase the chance of ADAS knowledge and usage, it is crucial to create methods to educate drivers on the use of ADAS. Harms et al. (2020) found that there are numerous different ways that drivers acquire ADAS knowledge. The most common way they found was thorough trial-and-error (47 %). This can be considered as an attempt to seek out the knowledge by oneself, in contrast to being shown by someone else. The second and third most common way was through reading a brochure (36 %) or instruction manual (27%). A study by Abraham, Reimer, Seppelt, Fitzgerald, & Coughlin (2017) found that 59 percent of car owners have read the user’s

(19)

8 manual for their car at some point. According to Carrol and Rosson (1987), users with all different levels of knowledge do what they can to avoid reading user manuals. Even though reading the manual is a common method to acquire knowledge, it might not be the most desirable way to find the information the user is looking for.

In the previously mentioned study by Harms et al. (2020), 24 % of drivers reported that they had learned about their ADAS functions at a car dealer. This type of demonstration-based learning can promote the opportunity to observe, but also perform the task in question. Performing the task can facilitate reinforcement learning for the specific task (Harms et al., 2020). If drivers are given a demonstration of ADAS, they are four times more likely to rate ADAS as useful, compared to if they receive a learning protocol (Nylén et al. 2019). This difference might be since the drivers in the demonstration group can more effortlessly create a mental model of how the ADAS can operate and the limitations of the system (Nylén et al. 2019). However, it is not always the case that instruction is provided at a car dealer. Abraham et al. (2017) examined different car dealers in the US and found that drivers would like a more in-depth introduction from car dealers.

2.7 Acceptance

In the end, acceptance of a system is necessary for the driver to incorporating it in the driving behavior. Acceptance is according to Dillon (2001) defined as “demonstrable willingness within a user group to employ information technology for the tasks it is designed” (p.1). Instead of emphasizing the intention of use, the definition highlights the significance of actual acceptance behavior. The understanding of lack of acceptance is central to comprehend what evokes resistance in potential users. Previous positive experiences with the system can also generate a higher degree of acceptance. For example, if the person or a relative to the person was saved by an early version of ADAS, it is more likely that the customer would purchase a car with an ADAS later on (Planing, 2014).

If there is a lack of understanding and knowledge regarding new technology, there will be a risk of fear effects according to Nabih (1997). With fear effects, what is unfamiliar could be perceived as scarier and vice versa. In other words, if a person works in the tech industry they might be more likely to use tech devices. This hypothesis is aligned with the findings by Chtourou and Souiden (2010) who claim that the usefulness of new technology is reported as higher if there is a larger enjoyment involved when using the product. Therefore, one can hypothesize that if one chose to work in the tech industry one finds this to some extent enjoyable and thereby have a higher chance of finding systems such as ADAS useful. A meaningful purchase, like a car, is according to Planing (2014) linked to both self-image and emotions to a higher degree than more common everyday purchases. Since car purchases are relatively rare it is more important to many consumers that the purchase is aligned with their self-image. Another factor that potentially can evoke resistance is the perceived restriction of freedom of choice. If the driver feels like the car “takes over”, in an uncontrollable manner, this can lead to a feeling of resistance (Planing, 2014).

Assimilation bias is the tendency to apply what a person already knows when they enter an unfamiliar situation (Carrol & Rosson, 1987). This is a natural effect, but it also means that different people enter new situations with an unequal amount of knowledge. Namely, if a person works in the automotive industry and buys a new car with the latest high-tech functionalities the person will see these functionalities partly from the perspective that they brought with them from work. On the other hand, assimilation bias can lead to a transfer effect

(20)

9 that is too general. For example, if a person works as a pilot and believes that the unfamiliar functions of the car could be used similarly.

2.7.1 UTAUT model

The Unified Theory of Acceptance and Use of Technology (see figure 1) is an acceptance model by Venkatesh et al. (2003) that combines several recognized models of individual acceptance. The model shows what factors weigh in when a user is accepting or not accepting, and thereby using, new technology. According to Venkatesh et al. (2003), the relationship between the factors depends on age and gender, but the “Performance expectancy” is the most common determining factor of behavioral intention.

Figure 1

Model of the Unified Theory of Acceptance and Use of Technology

UTAUT was modified by Adell (2009) to better fit the factors that matter when a driver is accepting or not accepting a driver support system. The suggested modifications to the UTAUT model (see figure 2) was to add a dimension of emotional response, Affect, and Satisfaction. The second modification was a possibility to weigh the importance of the different constructs. Lastly, system reliability was included in the model. However, the last modification is not illustrated in the modified UTAUT model.

(21)

10

Figure 2

(22)

11

3 Method Study I

In this chapter the method of Study I is presented. This includes the methodological theory, collection of data, participants, analysis process, and discussion of methodology.

A qualitative research approach was chosen due to the explorative nature of Study I. A qualitative approach was believed to have a better chance to reveal the underlying factors and thereby allow for a more nuanced research scope. A qualitative approach was considered to be better suited for an explorative study where the goal of the results of the study is rather vague in comparison to a study where the purpose for instance is to validate a specific theory. More specifically the method has its foundation in constructionist grounded theory, which is presented in the following section.

3.1 Grounded Theory

Grounded Theory (GT) was selected as the methodological base of this study since it allows the empirical findings from the data collection to guide the direction of the study. GT is well suited for Study I since there is not a specific theory this study aims to test. The GT method was created during the 1960s by Barney Glaser and Anselm Strauss. They argued that the sociological methods available at the time were only concerned with preconceived logically deducted hypotheses (Glaser & Strauss, 1967). In contrast, GT was an inductive, comparing, and interactive method (Charmaz, 2006). GT was according to Glaser and Strauss (1967) a method to allow exchange between empirical data and analysis of data to enable theory development through coding. By developing theories from empirical findings, it can generate refined or new insights within the field in question (Glaser & Strauss, 1967). The empirical findings then constitute the foundation for a theoretical framework that can be calibrated through more empirical findings. The GT approach is both systematic yet flexible and guides the researcher through the data collection and analysis process. The method aims to find theories “grounded” in the extracted data. Due to the qualitative nature of the method, the purpose is according to Glaser and Strauss (1967) not to collect data for population representativeness but rather to collect data until theoretical saturation is obtained. The state of theoretical saturation is reached when every new sample of data aligns with your stated concepts or categories.

3.1.1 Different versions of GT

After developing GT, Glaser and Strauss separately evolved the theory in different directions. Glaser went on to pursue an objective ontological approach, where his interpretation was that there is a reality that exists independently of the agents in it, in this case, the researchers (Guvå & Hylander, 2003). Glaser’s version of GT is commonly referred to as Glaserian GT, but Glaser prefers the term Classic GT (Glaser & Holton, 2004). According to Glaser and Holton (2004), Classic GT is “Simply a set of integrated conceptual hypotheses systematically generated to produce an inductive theory about a substantive area” (p.3). Strauss went on to pursue his research with Juliet Corbin, their approach being more prescriptive than Glaser wished for GT. Glaser believed that Strauss and Corbin’s approach forced theory on data.

Another researcher who has influenced the development of GT is Kathy Charmaz. She calls her a contemporary form of GT, constructivist grounded theory (CGT). In CGT a new epistemological foundation was adopted (Charmaz, 2006). Charmaz’s theory highlight the flexibility of GT, as opposed to the more prescriptive path by Strauss and Corbin. Charmaz’s constructivist viewpoint puts the researcher and research process in a historical, cultural, and social context. The theory acknowledges the unavoidable subjectivity of the researcher. Due to

(23)

12 the subjectivity of the researcher, Charmaz believes that the researcher is a co-constructor of meaning in the research process.

Another central aspect where the interpretation of the researchers differs in GT is theoretical sensitivity. The concept of theoretical sensitivity addresses the subjectivity of the researcher. If a researcher already has a high level of theoretical insight into the research subject, it can prevent creativity when interpreting data (Corbin and Strauss, 2008). However, according to Corbin and Strauss (2008), a review of the current literature can be used to increase the researcher’s sensitivity, as long as there is a balance between seeing relevant data and preserving creativity. Glaser, on the other hand, argues, from an inductionist point of view, that a clear and accurate understanding of empirical data can only emerge if the researcher does not bring preconceived categories or questions to the research process (Glaser, 2019). From a CGT perspective, a literature review can be used in a data-sensitive and constructive manner, without forcing in on data (Charmaz, 2006).

3.2 Analytic process of Constructivist Grounded Theory

CGT was chosen as the guiding conceptual and analytic framework for Study I. The method was chosen due to its social constructive approach to interpret data, where theory is believed to emerge in the engagement of researcher and participant (Charmaz, 2006). CGT has according to Charmaz (2006) an interpretive view on research, which is the position of this study. In this study, it is the opinion of the author that the review of literature is needed to establish an adequate level of theoretical knowledge. Thereby, the chances of asking relevant questions during the interview increase together with identifying relevant categories of data. However, this study does not emerge from one certain theory regarding the usage of ADAS. Therefore, CGT is a suitable methodological approach interpretation of the researcher’s role and technical sensitivity.

The steps of analysis in CGT is according to Charmaz (2006) to gather rich data, coding, and memo-writing. The steps of analysis are presented below.

3.2.1 Gathering rich data

In Study I, the data is collected through semi-structured intensive interviews. An interview is according to Lofland and Lodland (1984) a directed conversation. An intensive interview is, according to Charmaz (2006), a form of research method that allows for an in-depth exploration of a specific topic and where the person being interviewed has relevant experience. The interview structure of intensive interviews ranges from semi-structured to a more loosely guided approach. The interview fosters eliciting the respondent's own experience and personal reflections. Intensive interviews and CGT go well together since they both are directed, but still, flexible research approaches.

3.2.2 Coding

Coding is the first step from transforming primary data towards creating a theory from the grounded data. According to Charmaz (2006) coding is about categorizing data with suitable summarizing names. By coding the researchers begins to define the segments of data and start to understand the meaning behind it. The coding process begins with an initial coding phase. During this phase, the researcher remains open for all possible interpretations of the collected data. The coding, depending on the data, can be conducted word-by-word, line-by-line, or incident-by-incident. The choice of strategy depends on the purpose of the research and the level of abstraction. After the initial coding, it is time for the second phase, the focused coding. During focused coding, the researcher uses the found segments from the initial coding phase

(24)

13 and apply them to a larger amount of data. The entire coding process is iterative and constant comparison is a crucial part. In constant comparison the researcher codes and analyses the data in parallel with the data collection. The comparison starts at a lower level where the analyst compares data to data, e.g., to different interviews. Then the analyst can group similar data into concepts and compare that concept to other concepts. When comparing concepts to concepts, new categories arise. This process facilitates the researcher’s ability to separate concepts or categories from each other and promotes analysis on a more abstract level (Corbin & Strauss, 2008). From categories, properties of the concepts can be derived. A property of a concept could, for instance, be different consequences or conditions. Finally, the aim is to create a core category that highlights the most important category for the new theory. To summarize, CGT analysis involves taking data apart, conceptualizing it, and developing concepts in terms of their properties and dimensions. All to determine what the parts tell us about the whole (Charmaz, 2006).

3.2.3 Memo-writing

During the research process, the researcher should take time to write memos with informal analytic notes. Memo-writing is important in GT since it allows for data analysis throughout the entire research process. The purpose of memos is to evolve categories, define how the categories are related, and identify gaps. Memos can give room for the researcher’s spontaneous insights or new ideas that arise during the process. Figure 3 below visualizes the iterative process of CGT.

Figure 3

(25)

14

3.3 Data collection

The following section presents the conducted pilot study, the prescreening survey, participants and ethical considerations

The drivers of the highest interest for the analysis were drivers who have a negative attitude or low usage of ADAS. However, it was hypothesized that the negative attitude and low usage group could be hard to recruit since a driver with a negative attitude might not be willing to spend extra money on a system that they are hesitant about. However, if one part of a pair has a positive view on ADAS it might be the case that they buy a vehicle with ADAS. If both parties of the pair use the vehicle with ADAS, this would put the low usage driver in a situation where exposure to ADAS is inevitable. Therefore, pairs with access to a vehicle, but who had a difference of opinion regarding ADAS were recruited to Study I. Even though the low usage group was of most interest, the inclusion of the high usage drivers allowed for exploring the differences of usage and attitude between a high and low usage group.

3.3.1 Pilot study

A pilot study for study I was performed with one pair. After conducting the two interviews, one question about the partner’s ADAS usage was added and one scenario was changed. The question regarding the partner’s ADAS usage was added due to the discovery that it made it easier for the participant to put their own driving style in perspective. The previous scenario from the pilot study concerned eco-driving and though as a way to make the participant elaborate on how they reason regarding planning when driving. However, the participants were not familiar enough with the concept and the scenario was exchanged. Instead, the new scenario regarded the drivers' thoughts on driving in city traffic. Except for the two mentioned alterations, the structure of the interview was found to be efficacious.

3.3.2 Prescreening Survey

To create a balanced sample for Study I, a prescreening survey was distributed to multiple Veoneer offices in Sweden. It was expected that since Veoneer is a company within the automotive industry, there would be employees at Veoneer that had an interest in car technology. Therefore, they were considered more likely to have a car with the ADAS of interest in the study. To be included in the study it was required that they had someone in their household who also was suitable and would like to participate. Additionally, to be considered as a suitable participant they needed to have access to a car with ADAS, have a driver’s license to drive a car, and live in Sweden.

In the prescreening survey, the potential participants were asked demographic questions, such as about their age and gender. The potential respondents were asked how often they drove, if they had a driver’s license for any other type of vehicle than a car. They were also asked about the brand and model of the car they most frequently drove. Furthermore, they were asked if the car they most frequently drove had ADAS, and how often they would use the systems. It was requested that they filled out their attitude towards Driver support systems, automated vehicles, and manual driving. Moreover, they were asked to report how willing they were to take risks in traffic. Finally, the potential participants were asked to reflect on how quickly they adapt to new technology compared to others. The results of the survey were in addition to creating a balanced participant sample, used as support of discussion during the interviews. If two respondents of the prescreening survey were found suitable for Study I, they received an e-mail with a link to a new survey where they could select an interview timeslot.

(26)

15

3.3.3 Participants and ethical considerations

Out of the responses to the prescreening survey, five pairs who fit the participant criteria were selected for interviews (N=10). The participants consisted of six women and four men, aged between 21-60 (M = 49.4, SD = 10.78). There were four romantically involved pairs and the final pair consisted of a parent and a child. The five pairs were labeled with a number between one and five. Each participant within the pair was also given a letter to be able to report their responses in an individual, but anonymous manner. For instance, when a response is reported from the fourth pair the identification number is either 4A or 4B. This system is arbitrary and established to be able to highlight the possible pervading factors in a participant’s attitude towards ADAS. In the analysis, all participants are referred to as being a part of a pair. The term pair was chosen instead of couple since the term couple could be considered to only apply to two people who are romantically involved. Furthermore, in the results section the gender-neutral pronoun “they” was used instead of he or she. This was an effort to ensure the anonymity of the participants.

Before the start of the interview, the participants were all informed about the purpose of the study. They were also informed that the results of the study would not be traceable to them as individuals. Also, they were informed that their participation was voluntary and that they had the choice to end the interview at any given point if they wanted to. Lastly, the participants were informed that the interview would be audio and video recorded. The participants did not receive payment in exchange for their participation.

3.4 Data analysis of Study I

All interviews were conducted via the communication platform Teams or through a telephone call. Face to face interviews would be considered ideal under normal circumstances but due to the covid-19 pandemic it was not possible. The interviews were one on one, in other words not with both parts of the pair at the same time. The interviews were held separately as an effort to give the participants a chance to discuss their driving and their partners driving more openly. The interviews were recorded to allow for transcription.

When the time for the interview came the researcher reached out to the participant through a video call. The participants were encouraged to use their web camera during the interview, but it was not a requirement. Before the start of the interview, the participants were informed about the purpose of the study, that they would be anonymous, that the interview would be recorded and that their participation was voluntary.

During the interview, the participants were asked about their knowledge and use of ADAS. They were also asked to describe themselves as a driver and how their driving style and their partner’s, parent’s or child’s driving style differed. Related to the prescreening survey, they were asked to elaborate on why they had rated themselves at a certain risk-taking level. Their general interest in technology was also a topic of the interview. Finally, the learning process of ADAS was brought up. The participants shared their views on how they believed that the learning process of ADAS should look like. The exact questions of the interview are attached in appendix A.

The time the interviews lasted ranged between 18 – 43 minutes (M = 31.2, SD = 7.65). All recordings were transcribed. Afterward, segments of the interviews were categorized as incident-by-incident. The segments were then structured participant by participant. The structure of the coding analysis can be seen in figure 4. Each row represents a participant and

(27)

16 each column represents an incident (For instance learning how to use ADAS or describe your driving style). On the right side are examples of the memos written during the analysis process.

Figure 4

Coding process of Study I

The purpose of the data analysis was to discover patterns in the underlying reasons for none or low usage in ADAS. Primarily patterns of the ADAS the driver must activate themselves for usage, for example the ACC. Furthermore, the purpose was to differentiate the possible user groups and their characteristics. The drivers of the highest interest were the low usage group. To be categorized as a member of the low usage group the driver had to use ADAS that require activation from the driver half of the times when driving or less.

(28)

17

4. Results Study I

In the interview results section, a background summary of all participants is presented. Then the high-frequency users are presented on a group level, while the participants who belong to the low usage group are presented individually. The reason behind this is the focus on low usage in this thesis, the emphasis will be on the low usage group.

4.1 Driving background results for all participants

The participants reported different driving habits. The number of days they drove per week ranged from one to seven, with an average of 4.33 (SD = 2.22). Four of the participants drove to work every or almost every day. Three drove to work a few times per month. The distance the participants drove per year ranged from 1200 – 22 500 kilometers with an average of 10 220 kilometers per year (SD = 7129.18). All participants reported that their everyday driving mostly consisted of driving in a city environment. However, a few times per year they did longer trips to a summer house, to a ski resort, or to relatives that live in a different part of Sweden. All these trips included more highway driving than their everyday driving pattern. One pair who lived in the countryside reported more country road driving than any of the other pairs. None of the participants stated heavy traffic as a reason behind not taking the car into the city center. However, they mentioned other factors such as expensive parking, taking too much time, and environmental aspects. All participants reported that they enjoyed driving but to different extents. None of the participants stated that they take pleasure rides without a declared goal.

The results of the interview show that all respondents used ADAS every time they drove, even though three of them responded that they never or rarely used ADAS in the prescreening survey. The underlying explanation behind this gap seems to be that some respondents only thought to include the ADAS that they had to activate themselves when answering the prescreening survey, such as the ACC.

With the exception of a few not being familiar with Lane Keep Assist or Pilot Assist, all participants had heard about all the features included in the interview protocol (Rearview camera and sensor, adaptive cruise control, forward collision warning, automatic emergency brake, lane keeping assist, lane departure warning, blindspot warning, pilot assist).

4.2 Results high usage group

Six out of the ten participants (1A, 2A, 3A, 4A, 4B, 5A) were considered to belong to the high usage group. Their responses are thematically reported.

The respondents of the high usage group drove on average 4.8 days per week (SD = 2.4) and 10 900 kilometers per year (SD = 5389.8). When asked about the division of utility and pleasure regarding driving two of the respondents reported that they used to enjoy driving more when they were younger. Still, they all reported they still enjoy driving, especially in new surroundings.

4.2.1 Driving style

When asked to describe their role as a driver, four of the high usage participants explained that they see their ability to read and predict traffic situations as a part of what defines them as a driver. One of the participants stated that the most important thing when driving a further distance was to plan the driving so they would not disturb other drivers. It was also important to drive smooth, so their children do not get motion sickness. Another driver described themself as “Fast but safe” [Snabb men säker].

(29)

18 Five of the high usage participants did not believe that their driving style changes when their partner is in the passenger seat. The sixth high usage participant reported that they drive more carefully when their partner is in the car with them. Both through adjusting their speed to the legal, or below the legal, limit. They are also more careful with their ADAS usage when their partner is in the car, “When I have my partner with me, I might turn it (ACC) off when I am approaching a roundabout, or when I know it might do some strange maneuver because it loses track of the object in front of it.” [När jag har min partner med så kanske jag kopplar ur (ACC) då när jag kommer till en rondell och vet att den kan göra någon konstig manöver för att jag vet att den tappar objektet framför].

When asked to describe their driving behavior five out of six respondents in the high usage group reported that they sometimes drive over the legal speed limit. One of the participants explained that they use the ACC to make sure that they stay a maximum of 5 km/h above the speed limit. None of the participants in the high usage group reported fear of overtaking other vehicles. However, two of them stress the importance of only using the left lane when actually overtaking. They both dislike when drivers stay in the left lane for a longer period.

4.2.2 Technology

All of the high usage participants worked in the automotive industry. They explained that they like technology and learning how to use new technology. When adopting new technology, four from the high usage group reported that they adopt new technology “Earlier than most others”. A common factor in the high usage group is that they state that they are curious and like to learn how to use new technology through trial and error.

4.2.3 ADAS usage

The participants were asked about what they believed is the purpose of ADAS. One of them proposed that the reason behind ADAS is to create better predictability of other drivers. If cars for instance has a standardized distance to other vehicles, it will be easier for drivers to predict how the surrounding cars will act. Another participant believes that even though safety is one of the purposes, one reason behind it is monetary. They state that the goal of ADAS development is to produce vehicles that are fully automated and therefore the companies do not have to pay salary to a human driver.

4.2.4 Activation of ADAS

In this section, the ADAS that require activation when driving, as opposed to a default ADAS, are presented.

4.2.4.1 Adaptive Cruise Control

All of the high usage respondents reported that they use the ACC but to different extents. One person reports that ACC is the ADAS system that they use most often because “It works for the most part. That is what is good about it” [Den funkar för det mesta. Det är det som är bra med den]. The same participant likes to use it in city traffic including when there is a traffic jam. According to four other high usage participants, the ACC is great, but they see no point in using it when driving in a city environment with roundabouts and traffic lights. They believe that the system is too slow to keep up with the traffic flow. One participant uses the ACC 95-98 % of the time when they are driving. To them, the ACC is a way to make sure that they are staying around the legal speed limit. The last participant of the high usage group does not always use ACC on the highway because it takes away parts of what makes driving fun to them. They only use it when they feel a bit sluggish or when they keep surpassing the speed limit by mistake.

(30)

19 4.2.4.2 Lane Departure Warning and Lane Keeping Assist

The participants had a difficult time keeping the Lane Departure Warning and the Lane Keeping Assist systems apart and therefore their answers regarding either of the systems will be presented in the same context. Both systems have the goal of reducing the risk of accidents due to the vehicles leaving their lane. However, if this happens the LDW will only warn the driver while the LKA will try to actively steer the vehicle back into the lane.

Two of the high usage participants have deactivated the system. One of them says: “I have actually deactivated it (LKA) because it was so annoying. Since it warns if I do not turn on the indicator when I leave my lane. So it kept telling me what to do, which I do not appreciate” [“Den har jag faktiskt kopplat ur för den var så irriterande. Eftersom den varnar om jag inte blinkar när jag ska lämna mitt körfält. Så den är väldigt uppfostrande vilket jag inte uppskattar”].

Another respondent believes that there is no real need to use the LKA if you are an attentive driver. However, one driver states the system can be useful if used on a highway. Three of the drivers of the high usage group state that the LKA works poorly when driving on country roads. Since they, when there are no other cars around, use both lanes to drive straighter when the road is curved. When they do this the system warns them too much. They do not perceive their behavior as a risky way of driving and dislike the extent of how much the LDW and LKA reacts. Another driver is skeptical towards LKA and states that the system can create dangerous situations if the LKA tries to steer back into a lane when the road is slippery. Another issue the same driver highlights with LDW and LKA is the confusion when there is road construction with temporary sideline markers. This can cause the system to still use the original lines as the reference point when it should follow the temporary sideline markers. They are hesitant about the purpose of the system. They state:

If you would reach a point where you drive off the road because you were not paying attention, then other systems should have noticed it earlier. Driver alert, driver monitoring. Personally, I am questioning the point of Lane Keeping Assist [Om du kommer till en situation där du kör av vägen för att du inte haft uppmärksamhet då borde andra system noterat det här tidigare. Driver alert, driver monitoring. Jag ifrågasätter personligen nyttan av Lane Keep Assist] 4.2.4.3 Blind Spot Monitoring

None of the participants state that they dislike BLIS. However, two respondents in the high usage group view taking the system for granted as a risk. One of them says, “You get used to it quickly so now I will barley turn my head if I drive in a car without it, I think. That is the disadvantage with this” [Vänjer man sig snabbt vid det så nu kommer jag knappt vrida på huvudet om jag kör i en bil utan tror jag. Det är ju nackdelen med detta]. Another respondent believes that BLIS is a valuable system but reports that they should use it more than they are currently doing.

4.2.4.4 Pilot assist

Two of the participants of the high usage group see the PA as a fun gadget but they do not use it more than for fun. One of them does not have PA in their car but likes to try and explore the system in their partner’s car. One of them does not have PA in their car but likes to try and explore the system in their partner’s car. Their partner, who also is in the high usage group, does not use the system. The partner (who doesn’t use the system) is uncertain if the system

(31)

20 exists in the car. Two other participants report that they do not fully trust the PA. One of them says that they do not see the point with the system. The only occasion they use it practically is when they have to do something for a few seconds, like unwrapping a sandwich. Other than that, they only use it for fun.

4.2.5 Default ADAS

This section presents the high usage participants' views on the ADAS that often are activated by default when starting the car.

4.2.5.1 Rearview camera

Four of the respondents appreciate the rearview camera or sensor in their car. Even though one of them acknowledges some limitations with the system “It can give a false sense of security sometimes, even if these sensors are really good. There might be branches, or whatever, blocking the sensors so you have to look behind you and not just back up” [Det kan ge en falsk säkerhet ibland, även om de är väldigt bra de här sensorerna. Det kan ju sticka upp grenar eller vad det nu kan vara, för sensorerna så man måste kolla så man inte bara backar in]. Another respondent praises the “See in the dark”-feature included in their rearview camera. To them, it is the biggest gain the system brings. They admit that they thought the rearview camera was silly at first. That it was only meant for the elderly who have a stiff neck. However, when they started using the system themself, they changed their opinion. Another participant dislikes the system “I think everything is pretty crappy. I do not trust it. Not even the rearview camera that is an old system. If there is a little snow it will not work.” [Jag tycker att allt är ganska kasst. Jag litar inte på det. Inte ens backkameran som är en så gammal funktion. Kommer det lite snö så funkar det inte]. Nonetheless, they identify that the rearview camera probably is an outstanding system if you are driving a truck with very limited possibilities to see what is going on behind you

.

4.2.5.2 AEB and FCW

The results of the FCW and AEB will be presented together due to the participants' tendency to discuss them in the same context.

On two separate occasions, one participant has experienced that the AEB has kicked in. One of the instances was due to a car that drove onto the road where they were going. They are not certain whether it would have been a collision if the AEB had not activated, but they felt a great appreciation for the system when it happened. In their experience, there have been no false positives with the AEB in their car. Two of the participants have only received a warning from the FCW, but the system has never intervened. One of them likes to experiment with the system to see how it reacts when they position the car at different angles.

4.2.6 Reasons behind ADAS usage

Since all the members of the high usage group worked within the automotive industry, some of them had their first encounter with ADAS through work. Buying a new car with ADAS was the cause of the first contact for the remaining participants. When deciding to use ADAS, curiosity was one of the driving forces for all but one participant of the high usage group. The will to understand how the technology works was a contributing factor for the same group of five. One of them states “It did not take many minutes before everything was activated because

Examining factors for low use behavior of Advanced Driving Assistance Systems