Visual Communication Console: Sharing Safety-Critical Information between Heavy Vehicles and Vulnerable Road Users

(1)

Master of Science in Mechanical Engineering June 2019

Visual Communication Console

Sharing Safety-Critical Information between Heavy Vehicles and Vulnerable Road Users

Erik Lindström Dastan Gomli

Faculty of Engineering, Blekinge Institute of Technology, 371 79 Karlskrona, Sweden

(2)

This thesis is submitted to the Faculty of Engineering at Blekinge Institute of Technology in par- tial fulfilment of the requirements for the degree of Master of Science in Mechanical Engineering.

The thesis is equivalent to 20 weeks of full time studies.

The authors declare that they are the sole authors of this thesis and that they have not used any sources other than those listed in the bibliography and identified as references. They further declare that they have not submitted this thesis at any other institution to obtain a degree.

Contact Information:

Author(s):

Erik Lindström

E-mail: erlb14@student.bth.se Dastan Gomli

E-mail: dagb12@student.bth.se

University advisor:

Program Director of Mechanical Engineering, Christian Johansson Askling Department of Mechanical Engineering

Faculty of Engineering Internet : www.bth.se

Blekinge Institute of Technology Phone : +46 455 38 50 00 SE–371 79 Karlskrona, Sweden Fax : +46 455 38 50 57

(3)

Abstract

Background. Over the years, between 2013 and 2017, accidents between Heavy Goods Vehicles and Pedestrians have come to increase. Leading causes stem from inattentiveness and lack of communication between driver and pedestrians. With the advent of Autonomous vehicles, set to be able to reduce accidents, uncertainties in how communication and trust between humans and machines will be formed remains.

Objectives. The research aim has been to understand the difficulties and problems surrounding heavy vehicles, and the problems that today’s heavy vehicle operators faces, from which a technical solution that addressees the uncovered needs, is developed.

Methods. Design Research Methodology and MSPI Innovation Process has been used in combination for acquiring and validating information around the problem.

Shadowing sessions, unstructured interviews has been used for acquiring information.

Literature reviews have been done to find academic validation in hypotheses stated throughout the research. From the information gathered, iterative prototypes have been built.

Results. From the needfinding that was conducted, safety around trucks was the field on which the scope of the research was focused around. Due the larger size of truck, decision-making through eye contact and intention determining is set to be harder when dealing with heavy vehicles, leading to an uncertainty around heavy vehicles residing with pedestrians in how to act around these. With the operators of these vehicles finding the unpredictable nature of pedestrians and cyclist in traffic to be troublesome and safety imposing, the research aim was set around addressing these needs. A communication console was developed, that is able to communicate safety-critical information between heavy vehicle operators and vulnerable road users, as means of reducing front collisions between said parts.

Conclusions. The console has been developed through iterative prototyping and testing, with design requirements being acquired through learnings and feedback gathered from each iteration. The resulting communication console being presented is able to share critical information being sought by pedestrians for decision-making, primarily that of eye contact and intentions of oncoming vehicles. The system serves as a proof of concept, that could through extensive traffic safety testing, help reduce front collisions between Heavy Goods Vehicles and Vulnerable Road Users, as well as, through further development, become the central communication console for autonomous vehicles to ensure partnership and intuitive communication between these and the surroundings.

Keywords: Visual Communication, Communication Platform, Heavy Vehicles, Ma- chine Vision, Traffic Safety

i

(4)

(5)

Sammanfattning

Bakgrund. Från år 2013 och 2017 har antalet olyckor mellan lastbilar och fot- gängare ökat. Bakomliggande orsaker härstammar från distraktion och kommunika- tionsbrist mellan förare och fotgängare. Med tillkomsten av autonoma fordon, som kan möjligöra reducering av olycksantalen, kvarstår det den en osäkerhet i hur kom- munikationssätten mellan människa och maskin kommer se ut.

Syfte. Forskningens fokus har varit att förstå svårigheterna och problematiken kring tunga fordon, samt att förstå problemen som dagens tunga fordons förare upplever.

Från behoven har sedan en teknisk lösning utvecklats för att adressera dessa.

Metod. Design Research Methodology och MSPI Innovation Process har i kombina- tion med varandra för insamling och validering av information kring problematiken.

Shadowing och ostrukturerade intervjuer har använts för information insamling. Lit- teraturundersökningar har använts för att hitta akademisk stöd för dem hypoteser som formats under forskningen. Från informationen har sedan iterativa prototyper byggts.

Resultat. Från den behovsidentifiering som genomförts framgick säkerhet kring lastbilar som det fokus som forskningen valdes att fokuseras kring. På grund av tunga fordons större storlek blir ögonkontakt och intentionsbestämning svåvare för att genomföra för oskyddade trafikanter. Detta resulterar i osäkerheter kring tunga fordon och hur man som fotgängare även ska aggera kring dessa. Fordonsförarna finner oberäknerligheten hos fotgängare och cyklister att vara ett problem för säk- erheten och forskningen fokuserades kring att addressera och dessa behov. En kom- munikations konsol utvecklades som möjliggör kommunicering av den säkerhetskri- tiska information som söks och krävs mellan förare av tunga fordon och oskyddade trafikanter, med avsikten att reducera front kollitions olyckor mellan nämnda parter.

Slutsatser. Konsolen har utvecklats genom iterativa prototypskapande och tes- tande, från vilka designkrav har formats från varje iteration. Den resulterande kom- munikationskonsolen som presenteras delar kritisk information som fotgängare idag söker för beslutsfattande i trafiken, primärt ögonkontakt och intentionsbedömning.

Systemet tjänar som ett konceptbevis, som genom vidare trafiksäkerhetstester kan hjälpa reducera frontkollisioner mellan lastbilar och oskyddade traffikanter, så väl som att bli en central kommunikationskonsol för autonoma fordon i säkerställande av partnerskap mellan dessa och omgivningen.

Nyckelord: Visuell Kommunikation, Kommunikationsplatform, Tunga Fordon, Ma- chine Vision, Trafiksäkerhet

iii

(6)

(7)

Acknowledgments

A special thanks goes out to our supervisor Christian Johansson Askling, and Ryan Ruvald for the continuous support throughout the research, always taking the time of to help out, providing guidance, and clarification in times of confusion and struggle.

We would like to thank Thomas Lennartsson and Peter Blaschke for providing us with tools, material, guidance and support. Thanks to Omsri Addula for the help and development of the machine vision system on which the final prototype relied on.We would like to thank Jenny Elfsberg, Volvo Group Connected Solutions, Tobias Larsson, and Blekinge Institute of Technology for making this thesis possible.

We would also wanna thank all the those who have helped us throughout the research with giving feedback and helpful comments, and to the companies that provided us with resources and staff to conduct our research, providing us with shadowing possibilities by means of ride-alongs, as well as interviews possibilities.

Without you this would not have been possible. Thanks!

v

(8)

(9)

List of Figures

1 Pedestrian Decision Making Factors, from [29] . . . 8

2 Presentations of the traditional turning signal [28] . . . 9

3 Schematic Presentations of the brake light from 1930 [23] . . . 9

4 Schematic presentations of the first modern turning signal from 1926 [23] . . . 10

5 ÆVITA, from [28] . . . 10

6 Mercedes Front Zebra Crosswalk Projection, from [22] . . . 11

7 Mercedes Rear Text Display, from [22] . . . 11

8 FORD Lightbased Visual Language, from [9] . . . 12

9 Ergonomy Image Interpretation, from [26] . . . 13

10 TOBII Automotive Eye Tracking Concept, from [37] . . . 15

11 DRM Framework, from [20] . . . 17

12 DRM Relation, from [20] . . . 18

13 Validated Learning Cycle, from [30] . . . 28

14 Needfinding; Ridealong with garbage truck in Karlskrona . . . 32

15 Picture from ride-along with public transportation bus in Denmark, showcasing scenario of one bus having to make way for another. . . . 32

16 Blind Spot Mirrors, picture taken from Ride-along with delivery truck in Skåne, Sweden . . . 34

17 Truck test rig setup . . . 37

18 First iteration of eyesight projection . . . 38

19 Eyesight projection prototype close-up . . . 39

20 First iteration of intention prototype . . . 40

21 Headlight prototype pattern in sequence . . . 40

22 Second iteration - LED-strip, cardboard sign and headlight pattern . 42 23 Second iteration - combined prototypes . . . 43

24 Third Iteration . . . 45

25 Paper diffusion test on third iteration . . . 45

26 Fourth iteration, first pattern . . . 47

27 Fourth iteration, first pattern with paper diffusion . . . 47

28 Fourth iteration, second pattern . . . 48

29 Fourth iteration, second pattern viewed from 85 meters away. . . 49

30 Fourth iteration, second pattern viewed from 85 meters away, picture upscaled 1000% . . . 50

ix

(12)

31 Sequence of Machine vision system detecting a person coming into frame of the active camera feed, being able to determining positioning,

be it right or left. . . 54

32 Final prototype . . . 55

33 Final prototype mounted on a truck . . . 56

34 Eyesight projection . . . 56

35 Confirmation signal animation in sequence . . . 57

36 Speedometer deceleration animation in sequence . . . 58

37 Head Tracker cap that was constructed using the MPU-6050 as seen on the right . . . 59

38 Yaw, Pitch, Roll directions, from [18] . . . 59

39 Wiring Schematic of Final Prototype . . . 83

40 CAD drawing of dimensions for fourth and final prototype . . . 84

x

(13)

List of Tables

1 DRM Research Types, from [20] . . . 19

2 Learnings from second iteration . . . 41

3 Learnings from second iteration . . . 44

4 Learnings from third iteration . . . 46

5 Learnings from fourth iteration . . . 50

6 Design Requirements . . . 51

7 System Function Flow Diagram . . . 52

xi

(14)

(15)

Nomenclature

ADC Analog-to-Digital

ART Accident Research Team CE Construction Equipment DRM Design Research Methodology DS Desciptive Study

DSP Digital Signal Processing HGVs Heavy Goods Vehicles HV Heavy Vehicle

IR Infrared

MVP Minimum Viable Product PS Prescriptive Study

UV Ultraviolet

VRUs Vulnerable Roads Users

1

(16)

(17)

Chapter 1 Introduction

Every year there are approximately 26.000 fatalities on the EU roads and 15% of these are related to Heavy Goods Vehicles (HGVs) [33]. The causes of these accidents are primarly put into three categories: Human factors, Environmental and Vehicle, where human factor accounts for 90% of the caused accidents [33]. For this thesis project, the scenarios considered are the accidents between HGVs and Vulnerable Roads Users (VRUs), which consists of pedestrians, cyclist, moped riders and motorcyclists, but will primarily focus on pedestrians. According to Volvo Trucks Accident Research Team (ART) study, in 2014 HGV-related fatalities totaled 3,863 whereas around 650 involved pedestrians. [33]

According to Kockum et al. [33] accidents with VRU even happens when the driver is fully focusing on the traffic due to the unpredictability of VRU since their movements can leave little room for drivers to react. One major growing problem is that pedestrians, even cyclists, fails to pay attention to the traffic situation due to distraction by their smartphones [33]. The ART study shows that 17% of all pedestrians are crossing roads without paying attention due to being on their smartphones.

Adding to this, a comparison from ART’s study shows that accidents between HGVs and VRUs causing fatalities or severe injuries has increased from 15-25% to 30-35%

between 2013 and 2017. Typical accident causes that are brought up mention inattentiveness in multiple scenarios described such as crossing roads in front of the Heavy Vehicle (HV) or when HV make a right turn [33].

“If all road users were fully focused on their primary task – to move safely and be aware of the traffic situation – the number of accidents would decrease." - Kockum et al. [33]

The statistics point towards there remaining a problem concerning safety between HGVs and VRUs as the number has increased over the years. And as of 2017, 30%

of the accident between HGVs and VRUs, most commonly pedestrians, are of this type, which is the case being tackled in this thesis project. [33]

1.1 Collaboration Partners

1.1.1 Volvo Group Connected Solutions

The research beginning presented in this thesis has been done in collaboration with Volvo Group Connected Solutions. Volvo Group houses a number of brands within

3

(18)

4 Chapter 1. Introduction the transportation industry with trucks, buses and construction equipment, and aims to drive prosperity through its transportation solution and is one of the leading manufacturers within its domain.

For Volvo, safety is a defining core value, striving to achieve its zero accidents vision across which is shared across the Volvo Group brand. Volvo Construction Equipment (CE) having released a Triple Zero vision which aims to keep accidents, emissions, and unplanned stops to a zero [40]. And likewise work exist within Trucks and Buses [41][39]

Seeing to the exponential growth of technology and the impact can come to have on society, changing the current role of humans and how lives will be lived, Volvo wants to explore how use of technology can be used in improving the lives of HV drivers and operators, with a focus on automation, electrification, and connectivity.

Volvo CE is through its 10x Efficiency, is through electrification working to reduce energy consumption [40]. Volvo Trucks is looking into how connectivity between vehicle, via so-called road-trains, can come to reduce accidents via a combination of expertise from professional drivers and automated vehicle systems [41]. And Volvo Buses foresees a revolution within traffic safety to lie in autonomy, with having autonomous buses [39].

1.1.2 Stanford University and ME310

The focus of the study has come to be developed from the initial project prompt given by Volvo Group Connected Solutions for the ME310 course of 18/19, in which the researchers participated in a global collaboration between team of Blekinge Institute of Technology and the team of Stanford University. The scope of the prompt was narrowed down leading to the field given in this thesis.

1.2 Aim

The aim of the research is to get a better understanding of the difficulties today’s heavy vehicle drivers face in the scope of operating within urban environments. The goal is to develop a technical solution that can aid the heavy vehicle drivers of today, improving traffic safety and help build a foundation step towards the autonomous vehicle future that is to come.

1.3 Research Questions

• How can safety-related/-critical communication be supported between heavy vehicle drivers and Vulnerable Road Users?

1.4 Delimitations

The suggested technical solution will be based primarily on the case of it being applicable for usage on trucks, as it has been the primary source of data through the research.

(19)

1.4. Delimitations 5 The use case is limited to cities, as the long-haul drivers face different types of problem in addition to those faced within the cities. All in order to get more focused research and a solution that could cater towards a larger a target group, although having not primarily focused on it through the extent of the research.

The focus of the research is also limited to what is capable within the knowledge base of mechanical engineering. Meaning that fields, such as User Experience and Psychology will not be focused on in developing in the suggested solution.

(20)

(21)

Chapter 2 Related Work

2.1 Current traffic situation

Today roads are well-equipped with traffic signals, signs and markings to provide explicit guidelines for operators in and around the roadways. There are also many legislations in traffic that are important for the safety of all road users. Still, there are decisions, such as determinations of obligation to weigh that end up in negotiation between vehicle drivers and pedestrians [4]. Pedestrians often depend on cues in the driver’s behavior and must seek eye contact, postures or gestures from the driver to decide. This means there is often the need for communication between vehicle drivers and pedestrians, which means that uncertainties arise in pedestrians while there is a human driver [4]. Achieving a safe interaction means that all road users have a similar understanding of the situation otherwise we will have a breakdown in the interaction and it is more likely that accidents occur [7]. According to [11] the most common causation factors in pedestrian accidents is misinterpretations.

2.2 Pedestrians’ decision-making behavior

From studies on factors influencing pedestrian decision-making behavior, the researchers of this master thesis, about heavy vehicles, found some factors relevant for this research. Rasouli A. and Tsotsos J. K. [29] did an investigation on pedestrian behavior by gathering literature from experts’ studies on pedestrian behavior, studies from the early 1950s to 2018, to identify characteristics that influence pedestrian’s behavior both with classical pedestrian-driver interaction and with pedestrian- autonomous vehicle interactions. The identified factors were put into two groups, the ones that directly relate to pedestrians, pedestrian factors, and environmental factors, and these contain several factor categories which have their sub-factors and is interconnected to each other, shown in figure 1. The pedestrian factors include Social Factors, Demographics, Abilities, State and characteristics. The environmental factors include Physical Context, Dynamic Factors and Traffic Characteristics [29]. From the traffic characteristics, Rasouli A. and Tsotsos J. K. [29] explain that the size of the vehicle influences the decision making of the pedestrian in two ways;

pedestrians tend to be more cautious when facing a larger vehicle and it impacts the pedestrian speed and distance estimation abilities. This was even experimented by Caird and Hancock, with 48 men and women where the conclusion was that the larger the vehicle the higher chance of underestimating the arrival time which affects the waiting time [4].

7

(22)

8 Chapter 2. Related Work

Figure 1: Pedestrian Decision Making Factors, from [29]

Going towards autonomous and driverless vehicles, it creates a social interaction void [29], since there will not be any communication for the pedestrians to rely on.

One major challenge that Habibovic A. et al. [1] mentions is finding a balance of what, when and how to communicate between human-machine interaction (HMI) and gaining public acceptance of autonomous vehicles [1].

2.3 Visual communication

Communication is part of our everyday life, we communicate in different ways, whereas the three main types of communication are verbal-communication (speak- ing), non-verbal communication (tone, body language, et cetera) and visual communication [35]. Visual communication is the transmission information using symbols and imagery [35], and is a common type of communication in traffic, with signs, direction indicators, emergency vehicle lights, et cetera. Vision is the sense that provides most of the data we receive and the sense that people depend on the most upon of all the other senses; hearing, smell, taste, and touch [6]. Visual communication through photographs, animation, charts, et cetera, is according to Agrawala [21] fundamental to the process of sharing information and is the most common type of communication being used in traffic as a driver and a pedestrian or cyclist. The

(23)

2.3. Visual communication 9 typical form of communication between vehicle drivers and VRUs are eye contact and gestures, for example, drivers at pedestrian crossing communicate with pedestrians so they can cross the road without anxiety [25].

Sucha [36] stated in his study of driver-pedestrian interaction the effects and importance of eye contact between pedestrian and driver. Sucha [36] also mentions that the driver’s form of communication with the pedestrian is besides eye contact, flashing lights or hand gesture to give the pedestrian confirmation that it is free to cross. Another factor that Sucha [36] mentions is that drivers speed up to force the pedestrian to stop and not cross the road.

Several ways of communication between drivers and pedestrians today have come to evolve naturally through the year of interactions. But how has the development of visual communication system been and how was it before these systems? If we go back to the early 1900s, turning and braking signals were done by putting the arm out [14]. Signaling left turn, your arm straight to left, signaling right turn, you turn your arm up at a right angle and for signaling stop, you turned your arm down at a right-hand angle [14], Figure 2.

Figure 2: Presentations of the traditional turning signal [28]

So gestures was the main communication methods between vehicle drivers since early ages of vehicles. The first signal light that was invented was the brake light, in 1905, Figure 3, but drivers still found it significant enough to use hand gestures [24].

1928 it became required in several states in the USA to have brake lights [24].

Figure 3: Schematic Presentations of the brake light from 1930 [23]

1907 an invention for turning signal came along but did not get implemented.

For many years after inventions for turning signals came along but did not catch car manufacturers interest, until 1925, Edgar A. Walz Jr., developed the first modern style turning signal as a aftermarket device [14], figure 4.

(24)

Figure 4: Schematic presentations of the first modern turning signal from 1926 [23]

Signals from the vehicle out to the surrounding has been the same since but with some development in their appearance, the light form, how they light up, as we know it today, brake lights gets brighter when braking and turn signal is blinking yellow light from behind and in the front.

But going towards autonomous vehicles, with driver less vehicles interacting in the city and the amount of people in the city we have today and the growth to come, there will be a need for changes in the communication system. Some companies has already started testing different type of suggested solutions within the field of visual communication for vehicles. As early as 2012 a group of researchers at Massachusetts Institute of Technology (MIT) built a bio mimetic interface for automated and electric vehicles called ÆVITA [28], Figure 5. An interface was built that was able to give reactions in different forms, physical and with lights which gave the vehicle “feelings”.

Figure 5: ÆVITA, from [28]

2015 Mercedes also joined this field of visual communication and showed a concept vehicle, Mercedes that could project a zebra crossing in front on the ground in front of the pedestrians and gave a sound signal to confirm it is free to cross, Figure 6.

Also, it gives texture messages to vehicles from behind, such as, “slow” if it is slowing down [22], Figure 7.

(25)

2.3. Visual communication 11

Figure 6: Mercedes Front Zebra Crosswalk Projection, from [22]

Figure 7: Mercedes Rear Text Display, from [22]

As late as this year, 2019, Ford has tested this area with a light-based visual language to find out what could help autonomous vehicles communicate with pedestrians in the future. The vehicle was equipped with a light bar on top of the vehicle that indicates what the vehicle is doing and what it is going to do next [9], Figure 8.

(26)

Figure 8: FORD Lightbased Visual Language, from [9]

This shows that people have made their own type of non-verbal communication which motivates the researchers of this thesis project to go further with the visual intent communication system in HV.

2.3.1 Ergonomics

Ergonomics is a broad discipline which has, according to Long J. [19] three domains contributing to comfort, satisfaction and safety. the three domains are the following:

Physical ergonomics (e.g. workplace design), cognitive ergonomics (e.g. how we perceive and process information) and organizational ergonomics (e.g. job design) [19]. Within ergonomics, there are many sub-specialties, one of them being visual ergonomics recognized by the International Ergonomics Association (IEA) formally in 2009 but wasn’t given an approved definition until 2012. The final definition defined by the Nordic Ergonomics Society and International Ergonomics Association Visual Ergonomics technical committee is:

“Visual ergonomics is the multidisciplinary science concerned with understanding human visual processes and the interactions between humans and other elements of a system. Visual ergonomics applies theories, knowledge, and methods to the design and assessment of systems, optimizing human well-being and overall system performance. Relevant topics include, among others: the visual environment, such as lighting; visually demanding work and other tasks; visual function and performance;

visual comfort and safety; optical corrections and other assistive tools.” - IEA [8]

2.3.2 Cognitive ergonomics

Cognitive ergonomics is important in the design of complex, high-tech or automated system [3]. It deals with mental processes such as perception, memory, reasoning and

(27)

2.3. Visual communication 13 motor response since they affect the interaction between human and other elements of systems [15]. For example, the user interface in a phone can cause frustration if it’s poorly designed. The aim is according to Kim, J Ergonomics (2016) to create appropriate communications amongst human needs, works, products, environments, capabilities, and limitations.

When people receive information, the mental process is distributed by attention [26]. Since the attention is limited, spreading the attention out will reduce the quality of the understanding. Osvalder & Ulfvengren (2010) states that attention can be directed towards what’s most important depending on stimuli qualities, experiences, and interests. Focusing on some stimuli and shutting out others is what attention is basically about.

Perception is about how the human gets aware of information surrounding. It stands for how the human thinks, understands the surrounding and remembers. This is a process that organizes the attentive stimuli and gives them a meaning depending on the context and here factors such as internal and external factors. Internal factors are the needs, experiences, feelings, and expectations. External factors are the size of the stimuli, contrast, insensitivity, and frequency. Figure 9 below is a picture of black fields with white background, but from experience and wishes to get an entirety of what is shown, a picture of a dog drinking water is visualized in the mind. [26]

Figure 9: Ergonomy Image Interpretation, from [26]

2.3.3 Color

Color is a powerful and effective way of communicating a message without words.

Even though color can have different meaning and differ from person to person depending on their interest in the color and even in different situations. Color is affected by culture, religion, age, light, environment etcetera [43]. One color can have a pos- itive and negative impact. Psychologist Angela Wright writes that there are four psychological primary colors; Red, yellow, green and blue that are often used for a different kind of communication [43]. These colors relate respectively to the body, the mind, the emotions and the essential balance between these three. The general

(28)

14 Chapter 2. Related Work classifications of colors are warm colors, cool colors and neutral colors [31].

Warm colors:

Red - Physical

Red is the color with the longest wavelength, makes it a powerful color. Because of its property of appearing to be nearer than it is, it grabs the attention first thereby effective in traffic lights. Wright A. 2004 also mentions the psychological effect stim- ulates and raises the pulse rate. Red has become a commanding color, which is why it is used for crucial information in traffic [31].

Yellow – Emotional

Psychologically, it is the strongest color because of its emotional stimulus. The right tone of yellow will represent confidence and optimism but too much of it will turn it around and represent fear and anxiety. Communicates temporarily and per- manent dangers which is why it is used in between red and green light, preparing the driver for a stop or to drive. Also used nearby construction in traffic or even on construction machines, cautioning for dangers [31]. The visibility of the color during day and night makes it an effective color in traffic for communicating important messages to road users [31].

Cool color:

Green – Balance

The green color is a restful color, the color of balance. It indicates the presence of green, because of plants and tree around the world grows near water and therefore indicates little danger of famine. In traffic, it is intentionally put out to not command the road users attention [31].

Blue – Intellectual

Wright A. describe blue as the color of the mind and soothing and thereby the world’s favorite color. While the red color has physical effects, the blue color affects mentally which makes it the color of clear communication. At the same time, it can be perceived a cold, unemotional and unfriendly [43], making it an effective color for police vehicles as warning lights, catching the attention of road users immediately.

Some countries have red and blue colors on the police cars, communicating command and warning [31].

2.4 Machine Vision

Machine vision is the ability for a computer to see, by converting analog-to-digital (ADC) data and digital signal processing (DSP) through one or more cameras [32].

Two important specifications in machine vision, that M. Rouse mentions, are the sensitivity and the resolution. The higher the sensitivity the better the machine vision detects weak impulses at invisible wavelengths. M. Rouse makes a comparison of the wavelengths with the human eye. A typical human eye responds to wavelengths from approximately 390 to 770 nanometers, while video camera can respond to much wider than that, some of them functioning at infrared (IR), ultraviolet (UV) or X-ray

(29)

2.5. Eye Tracking 15 wavelengths. The resolution is the machines ability to differentiate between objects.

Since these two factors are interdependent, increasing the sensitivity reduces the resolution and the other way around. Some example applications where machine vision is used can be medical image analysis, object recognition, pattern recognition, and electrical component analysis [32].

In urban traffic, pedestrians interact close with vehicles around them hence the safety of pedestrians is an important aspect. With machine vision, object and pattern recognition are two applications that can help the vehicle to identify VRUs and enhance their safety. for this to happen, D. Varytimidis, F. A. Fernandez, B. Duran and C. Englund (2018) [5] states that the most important feature of the machine vision is to first identify possible hazardous situations and secondly safely maneuver to avoid any collisions. Through visual communication systems in vehicles, such as lightning, methods of transmitting messages to other road-users are provided [25]. Through machine vision the vehicles can be enhanced to not only transmit messages to neighboring road-users but also, as Onishi, 2018, point out, to receive messages from neighboring road-users, such as recognizing eye contact and gestures from pedestrians, this is a fundamental feature for automated vehicles to have.

2.5 Eye Tracking

There is ongoing work within solving inattentiveness within the traffic. Tobii, a world-leading manufacturer within eye tracking technology is investing in solutions to deal with the manner by detecting distractions [37], Figure 10. Volvo Cars are in putting cameras inside newer upcoming Volvo’s to combat drunk driving and drowsiness, drowsiness which can be classed as a form of inattention. This system is working through machine vision and sensors to detect emotions and eyesight, which is useful for this thesis project to project the eyesight of the driver on the display for eye contact.

Figure 10: TOBII Automotive Eye Tracking Concept, from [37]

(30)

(31)

Chapter 3 Method

3.1 Design Research Methodology

The Design Research Methodology (DRM), figure 11, provided by Blessing & Chakrabarti is the main research methodology used for this master thesis to provide a framework for design research, help identify research areas, support research approach and methods and to provide guidelines for systematic planning of research. The methodology helps to achieve transparency and repeatability of the research being carried out. It also helps to increase the chances of obtaining valid and useful outcomes. Although outcomes of the research can still not be guaranteed since the nature of methodology is heuristic rather than algorithmic and what makes each research process unique is the personal interest and background. [20]

Figure 11: DRM Framework, from [20]

The framework for DRM contains four stages: Research Clarification, Descriptive Study I (DS-I), Prescriptive Study (PS) and Descriptive Study II (DS-II). Each stage

17

(32)

18 Chapter 3. Method is described further in the oncoming chapter below. Figure 11, shows how these stages are linked, what the basic means are and what the main outcomes are supposed to be. Each step of the framework shows the research methodology but the approach for each step is through the design thinking approach [12]. DRM and design thinking go hand in hand since they both mainly follow the same stages where DRM covers the research part and the design thinking covers the approach for the design. Figure 12 shows the relationship between DRM, design research and design to give a better understanding of how DRM and design thinking go together where design is where the design thinking process is implemented.

Figure 12: DRM Relation, from [20]

Relationships between design, design research and design research methodology DRM provides 7 types of design research, tabel 1, that is based on whether a particular stage requires initial study, comprehensive or whether a review-based study is sufficient. The choice of these types can be because of time restriction or because the project is part of a larger program which limit your priorities to in-depth study at each step. The review-based study bases the study entirely on literature, comprehensive goes deeper into the study by including literature and results produced by the researcher through empirical data gathering and developing support for it.

An Initial study covers all the steps in a particular stage but in less detail and only focuses on application evaluation, thereby closes the project and prepares the results of the project for others to use.

This thesis falls into the 5th type of design research since it is the most suitable for this type of project because of the research question and the available time frame and resources. In this project, the researchers identify support in literature reviews to clear out any prior assumptions coming into the research. This to get a clear goal and focused research, based on prior academic work. Thereafter analyzes the existing situation on a deeper level and goes beyond literature and starts observing and interviewing in relevant people in the relevant area. After addressing potential

(33)

3.1. Design Research Methodology 19 Table 1: DRM Research Types, from [20]

problem definitions that can be corrected and improved the researchers goes into the innovation and design thinking process to develop a solution for the addressed problem through iterations of prototyping. Since the project has a short limit time frame and resources the final solution is a prototype and concept of what could be a solution for the addressed problem, therefore it is closed off by an initial description of what’s needed for further development to reach the desired situation.

3.1.1 Research Clarification

The aim of this stage is for the researchers to identify evidence or indications and measurable criteria that support the assumptions or hypothesis that’s been made to reach the goal. The reason for this is to be able to formulate and execute a worthwhile and realistic academic and practical research goal. According to Blessing and Chakrabarti, (2009) the research clarification stage is mainly based on literature review, but this does not fit this project since the prompt was wide and there is poor research on the given prompt. Literature did not cover everything hence getting practical knowledge which was gathered through observations and interviews out on the field of research. By doing this a better understanding of the area of interest is reached and criteria were built to set a goal. Since the stages of DRM or design thinking is not a rigid and linear process, iterations are done several times on the stages and the goal will be refined and changed along the way of the process when the research gets deeper and knowledge is increased.

(34)

20 Chapter 3. Method

3.1.2 Descriptive Study I (DS-I)

In the first descriptive study, the researcher gets more detailed and go deeper into the literature to locate influencing factors to elaborate on the initial description of the current situation. This stage provides a better understanding of the current situation and offers clearer and more detailed descriptions of the current situation and to find success criteria. However, literature is not enough to find such detailed factors, which means the researchers will need to collect more empirical data through observations and interviewing stakeholders to obtain a better understanding and to determine such crucial factors. The next stage (PS) is then based on the finding from this stage which means that this stage should be detailed and well described.

Also, since these stages are not linear, the researchers find themselves iterating back to the RC stage and do some more literature review.

For this thesis, the researchers were able to gain a clear view of the current situation after four main iterations back to the RC stage. The researchers were able to map out crucial factors and critical point to aim through observations and interviews.

3.1.3 Prescriptive study (PS)

In the third stage, Prescriptive Study (PS), the researchers start to develop design support for the desired situation by taking the findings and knowledge from the second stage DS-I and determine which key factor to be addressed in order improve their initial description of the desired situation. The support can be knowledge, guidelines, tools, et cetera, that is used to enhance the current situation and to eliminate or reduce the influence of the critical factors found in DS-I or DS-II. The key factors can then have several impacts on the current and desired situation and therefore they are put into a different situation and then the researchers can approach with the factor that they find to give the best outcome and closest to the desired situation. According to Blessing and Chakrabarti, (2009) development relies on single findings, on assumptions and sometimes on experience since there is little evidence of extensive use of valid empirical data. Therefore, creativity and imagination are of high value to develop effective and efficient design support. The PS is a design task itself where the design support or its concept as the end product should be able to be evaluated to be able to evaluate whether the design support meets the criteria for the desired purpose. In this master thesis the design support is in form of prototypes which are tested on real stakeholders in the actual environment but even in a created test environment, to get valuable feedback, and then analyzed and evaluated. In this way, the researchers could take the most successful concept that fulfills the criteria towards the desired situation the best and develops that concept further and prepares for DS-II.

3.1.4 Descriptive Study II (DS-II)

The final stage of the framework is focusing on evaluating the impact of the support against the initially set criteria. Here the researchers have to evaluate the applicability of the support and the usefulness. Through iterations of prototyping and

(35)

3.1. Design Research Methodology 21 testings, the researchers will be able to compare tests and improvements to further on explain and build knowledge on the evaluations to find effects that they had not thought about earlier.

This stage can not always be on the same level of evaluation quality, for two reasons according to Blessing and Chakrabarti. One is that the project duration does not allow a full evaluation and the second is that the support has not been developed to the extent that it can be used by future users as intended. But the researchers still want to be able to evaluate some of its applicability, usability, and usefulness. Since this master thesis falls into the 5th type of design research, the DS-II is initial because the project duration is short, which means on this stage the research is getting wrapped-up and prepared results for other researchers to use. On the initial level of DS-II, the minimal requirements are according to Blessing and Chakrabarti:

• an indication of the applicability, usability, and usefulness of the support;

• an indication of the issues, factors, and links that need detailed evaluation;

• a suggestion for a proper Evaluation Plan.

The researchers evaluated the design support concept with tests and analyzes and then drew conclusions about how the relationship between the support and the aims of the research project are linked.

(36)

3.2 MSPI Innovation Process

3.2.1 Initiation

Framing

The first stage of the innovation process is to set the stage for an effective process.

Here the researchers clarify the goal and expectations as a team, which is connected to the Research clarification from the research framework given by Blessing and Chakrabarti (2009). This stage contains Framing, Teaming, and Planning. Framing the project is done through “HOW MIGHT WE”-questions. Using “HMW” puts the researchers into a mindset that allows them to come up with impactful solutions by allowing the problem to be open and solution to be independent [12]. This means that the “HMW”- questions should not put the researchers in a situation where they do not know where to start, the question is too broad, neither should the question be too narrow, that limits the creativity. The focus here is on the user and their experience hence the questions should not lock or direct the solutions. These questions will give the project a push start and will, later on, evolve and change with more knowledge and experience.

Teaming

The teaming part is to let the group work towards a common goal as effective and efficient as possible by creating a contract and setting the goals, expectations, roles, collaboration, Communication, decision making and conflict resolution [12]. The researchers did this together in a meeting filling in an informal contract within the group to be on the same level and to get the project flow smoothly and increase the creativity and cooperation in the group.

Planning

Planning the project is important to set milestones and making the project go forward. Here the group sets the strategy for how to go forward and what to prioritize to get to the common goal. The planning for this thesis was, in the beginning, an overview of the important milestones that had to be reached, so it was not a detailed plan. It was updated through the project with weekly and even daily milestones to reach the main milestones with to-do-lists written on the board to have it visual and be reminded of what to do.

3.2.2 Inspiration

Needfinding

This stage is where the needfinding is done. This stage is to find motivation for the search of solutions for the project [2]. From the previous stage the project was framed, and a problem was stated and then to search for solutions, empirical data was gathered to collect more information about the problem in question and look for user needs. The point here is to look for needs and not solutions since needs last longer than solutions [12]. The researchers went out to the field of research and

(37)

3.2. MSPI Innovation Process 23 did observations, interviews and ride-along, where they got the opportunity to work with heavy vehicle drivers for a day and get to see and feel their pains and needs, to gain deep customer empathy which is about understanding the customer better than they understand themselves [10]. Interviews were even done with other stakeholders from outside the direct scope such as city planners to gain an understanding of how heavy vehicles are seen from their perspective in our future cities going towards smart cities.

Trendwatching

Developing a solution for a problem the researchers had to make sure that the solution is a long-term solution and does not get outdated in the near future. By looking at where the world is heading and identifying emerging market behavior and social trends, the researchers can develop long-term solutions and open up for future opportunities [12].

3.2.3 Ideation

Connected to the trends are the emerging technologies to understand the future of potential technologies and services. This stage was done initially to be on the right track of technologies when ideas were generated since this was not the focus of the project.

Divergence

After gaining insight and knowledge in the field of study its time to bring them into ideas. The first stage of ideation is to expand the range of opportunities through different methods such as brainwriting, brainstorming, object brainstorming, et cetera, but this does not have to be done in a methodological manner, it is mostly about finding your own way to ideate [12]. The researchers went through the gathered data and insight and generated ideas individually which was then discussed together. There- after they did brainstorm sessions together so that the researchers could influence each other with ideas they had not thought of. When the researchers felt they had covered the exploring ground they had gathered from the insights they could move on to narrowing down the ideas in the next stage. This was not a onetime activity since this process is an iterative process.

Convergence

After some iterative brainstorming sessions, it’s time to converge and make decisions to pick an idea to develop further and test. There are several ways to evaluate ideas and rank them, such as dot-voting, Pugh matrix and by making sure the 3 overall success criteria from design thinking are met: desirability, feasibility, and viability.

Which means that the ideas have to be useful - desirable, be possible - feasible and be profitable – viable.

(38)

3.2.4 Implementation

Prototyping

After diverging ideas and then gone through the convergence and chosen out ideas within the criteria that was set, it’s time to bring these ideas into life and start exploring them. At this stage, the researchers start to design concepts to test their ideas and interact with the customers. At this stage the concept is put in the form of prototypes which, according to Johansson and Larsson [12] comes in 3 different type:

Visual prototype – where the concept can be seen, tangible prototype – getting to feel concept and how it’s used and experience prototype - getting the whole experience of the concept, close to the final product. The prototype should answer the “what if?”- questions in the search for solutions and should, therefore, be built fast to be able to get feedback and be easy and cheap to make changes or even discarded if it’s not right.

Since this is in an early stage the researchers should embrace failure and therefore be ready for many iterations and explore different opportunities [12]. As Johansson and Larsson explain, the researchers should not get emotionally attached to the prototypes and should let the prototype stand for themselves instead of defending them. In that way, if the concept is the right it, it will convince the customer for itself, if it does not then it probably not the right solution. The researchers built several simple prototypes in different forms, went out to the stakeholders and other users and tested these to get feedback and increased knowledge to be able to develop to the right final product that covers all the criteria and doubts.

Pitching

Together with the prototypes, a pitch will make it more convenient and give the concept a stronger ground. Pitching an idea is a way to communicate the idea, telling how it works, why it’s a good solution and whom it benefits. Pitching an idea can be done through video or a simple presentation and can be aimed at different audiences through different types of stories [12].

(39)

3.3. Literature Research 25

3.3 Literature Research

The DRM framework helps with the methodology to conduct the needed documenta- tion correctly for the research to get the academic content [20]. Literature Research was done in parallel with the progress of the research work, but with increased insight into the research area, new knowledge emerged during the process which led to the research evolving and gave new areas to examine. Depending on what the researchers were investigating, research was reviewed to motivate and gain more knowledge within that field [20].

The gathered empirical data provided the research with information to investi- gate and the researchers could gain new knowledge. Having investigated previous Academic Reports, Papers and Articles to access whether there is a need for continuous work within the field. Benchmarking whether there is something available or being developed with the field that is being perused with the research work being presented in this report. It also made it possible to reference and make use of already completed studies, saving time and boosting the work of the research.

The research material of academic articles, reports, and papers was gathered from databases, such as Google Scholar, Scopus, Research Gate. A strategy for finding works that could contribute to the research was through keyword searches that, at the given time was relevant for the research. Relevance was determined by reading through the abstract, something that Blessing and Chakrabarti suggest when going through several works of literature. Having read the abstract, the conclusion and introduction were proceeded to be read, seeing if similarities in results or contradicting results were present, being compared to initial assumption the research had, or what testing and data gatherings had proved. The most common keywords that directed this thesis were: automation, connectivity, maneuverability, visual communication, road safety, machine vision. Popular science articles were primarily acquired through Google searches, using the same keywords as mentioned above. These articles served a purpose through Benchmarking and Trendwatching, seeing to the ongoing work within the transport industry.

(40)

3.4 Research Approach

According to Cohen L., Manion L. & Morrison K. [17] The purpose of the research determines the methodology and design of the research. Since this is an experimental project of an invention, experimental research is appropriate [17]. As mentioned earlier, the DRM and MSPI Innovation Process are similar in their methodology where the DRM framework provides with strategies to support the research with academic research [20] and the MSPI support concept and design development [12].

Therefore, the researchers found it suitable to combine these two methodologies and use the DRM framework as the strategy for the research approach.

3.4.1 Qualitative and quantitative research strategy

A quantitative research strategy is used to find out to what degree, in terms of quantity, amount, number, frequency, et cetera, a phenomenon occurs [13]. Meaning the researcher bases his understanding of a problem on theory and measuring it by means of variables and questions. This strategy is based on closed questions, which means that the question must be complete and define the problem by applying the problem definition to a checklist addressing 5 key criteria’s : [13]

• Researchable

• Relevant

• Informative

• Reliable

• Effective

Qualitative research strategy does not start of the research with theoretical notions but go out in the field and observe and sensitizing. This is used when theoretical knowledge is incomplete. As Jonker (2010) mentions, it’s important for this strategy that the researcher grasp it “from the inside out” instead of “from the outside in”.

This means that the researcher should integrate into the field and get to know the stakeholders by partaking in activities. In this way, the researchers can see from the stakeholder’s perspective and create a better understanding of the field of research.

In order to achieve a full and “pure” understanding of the stakeholder’s behavior, the researcher needs to be as unprejudiced as possible. The purpose of qualitative research strategy is to search for and develop a theory, where theoretical notions can lead to “mini-theories”, which is a theory that works for a particular situation and needs to be proven validity and can further on, through repetition, lead to a theory that is applicable for several situations. [13]

The researchers chose the qualitative research strategy for this research due to the, quite new, wide range of research. Therefore, the researches needed to understand the behavior and effects of heavy vehicles drivers and their surrounding by conducting knowledge through observations and interviews.

(41)

3.5. Data Gathering 27

3.5 Data Gathering

3.5.1 Shadowing

Shadowing was used during the initial needfinding in order to get a first-hand perspective of the day to day situations that the target users of Heavy Vehicle operators experience. Through having ridealongs with truck drivers within different profession areas, such as waste and scrap collection, and delivery. A wide range of insights and information was obtained, which elsewise would have been hard to acquire from auditable (i.e. interviews, conversations) or literal (i.e. literature reviews, internet searches) means. By following different professions groups increased the likelihood of a common and shared problem amongst truck drivers to be identified, by means of getting down to a core problem something which Johansson and Larsson suggest in [12]. During the shadowing, normal conversations were held throughout the obser- vation session. Questions about a certain topic and free discussions were held. The free form of talking and knowledge acquiring provided insights which might have not appeared through a normal questionnaire. Additionally, questions were asked at times when different tasks were carried out, as to understand the underlying purpose and motivation behind it [Studying Actions in Context: A Qualitative Shadowing Method for Organisational Research]. The free elaboration that the shadowing provided, allowed the answers to be wider and more honest opinions. The duration of the shadowing sessions was full work-day, as a way of enduring the most tedious part of the work hours, gives way for unspoken needs to be aired, which elsewise never would have been able to be acquired.

3.5.2 Unstructured Interviews

Another method used during the initial Needfinding that was that of Unstructured Interviews [34], which were used to gather a vast amount of information from knowl- edgeable people within the field of Heavy Vehicles. Such people consisting of Garbage collectors, people within Logistics, and even within the field of City Planning. The widespread of fields interviewed were done as part of diverging the problem space with getting input from as many possible sources that act as stakeholders within the field. With hearing to the different stakeholders’ thoughts gave the possibility of spotting a common denominator from which further research within the field could be done.

The reasoning for conducting Unstructured Interviews was to give the people questioned the possibility of talking more freely about the topic, seeing that struc- tured Interviews could possibly run the risk of narrowing down the scope, possibly leading the unspoken problems not being noticed. Another reason being that initially little was known within the different fields, and having yet to settle around a given problem, the unstructured interviews could lead the way of finding new paths to explore.

The topic of the discussions was set around the field of Heavy Vehicles, Autonomy, Electrification, and Connectivity, within the scope of the urban areas, from which the interviews took shape as the discussions progressed. Outlining topic questions were used as a mean of getting the conversation started. Notes were taken on what

(42)

28 Chapter 3. Method was considered in driving the research forward, finding new fields to research further.

3.6 Research Quality

To validate the iterations and concepts in this research, the researchers used the Build-Measure-Learn loop given by Ries [[30]], figure 13. This model is according to rise the way to go from an idea to a successful product by measuring the customers’

reactions, gather the feedback and with the acquired knowledge decide whether to pivot or persevere [30]. After choosing an idea to build, it is time to build a prototype of it. The prototypes are experiments done on the customer to gather learnings [30]. The learnings come in the form of qualitative and quantitative feedback. The qualitative being, what the customer liked and did not like about the prototype and the quantitative being how many people used it and found it valuable [30]. For this research the researchers had their focus on the qualitative feedback rather than the quantitative since the time limit of the project did not allow them to build a product to the market. This is an iterative loop which by every iteration gain more learnings to refine the final product, it is therefore fundamental that the total time through the loop is minimized [30].

Figure 13: Validated Learning Cycle, from [30]

The limited timeframe of the research required the researchers to use the iterative

(43)

3.6. Research Quality 29 process’ with short timeframes, which made this approach fit well for this type of research. To minimize the time through the loop, the researchers built a quick version of the product that just showed the functions of the idea, this is called a minimum viable product (MVP) [30]. An MVP is meant to use the minimum amount of effort and least amount of development time and it lacks in many features, but it shows the desired functions and can gather feedback from the customers [30]. Prototypes were built in cardboard, aluminum, plastic, paper, glue, wood etc. The easiest way possible to be able to show the function. After every iteration the researchers went through the feedback and learnings and analyzed. If the prototype was to persevere, the researchers refined the prototype, added the needs of the customers and did another testing iteration. It the prototype was to be pivoted, then the researchers started another iteration of the MSPI Innovation Process, by going back to the inspiration and ideation stages.

(44)

(45)

Chapter 4 Results

4.1 Shadowing

Throughout the research a total of six ride-alongs were held within different profession groups, which consisted of the following fields and places from which the shadowing took place:

• Bus Public Transportation (Copenhagen, Denmark)

• Scrap Collection Truck (Kristianstad, Sweden)

• Waste Management (Karlskrona, Sweden)

• Waste Management (Kristianstad, Sweden)

• Delivery Truck (Skåne, Sweden)

From shadowing sessions that were conducted, a shared problem that drivers and operators of heavy vehicles shared was that of maneuverability. With operating such a substantially larger vehicle to that of a car, within urban cities which seemingly are not at all time considerate of heavy vehicles in the way city has been shaped, made it apparent that difficulties were faced. Narrow city streets with parked cars alongside the roads, which further posed difficulties, put a lot of stress on the driver to operate the vehicle in a safe manner.

With maneuverability standing as the most prominent issue faced by heavy vehicle drivers, a subsequent issue was seen, that of safety around Heavy Vehicles. The shadowing session of following the waste management truck in Karlskrona showed a case of eager road users wanting to push past the stationary garbage truck. The situation that arose showed a bicyclist zooming past the truck as garbage bin was being loaded in the rear loader of the truck. It was told that the situation is a common thing for drivers of garbage trucks.

31

(46)

32 Chapter 4. Results

Figure 14: Needfinding; Ridealong with garbage truck in Karlskrona

Figure 15: Picture from ride-along with public transportation bus in Denmark, showcasing scenario of one bus having to make way for another.

(47)

4.1. Shadowing 33 Further research on the matter showed that there is an issue present with how Heavy Vehicles and bicyclist are to coexist on the roads in the urban cities. The so- called blind spots present around trucks has had a history of being one of the causes of many accidents, due to the larger size of the vehicle makes the blind spots larger.

Within the EU legislation states that newer trucks are to be fitted with blind-spot mirrors to minimize blind spot, the legislation has been active since of 2007, and for older trucks since 2009 [16].

Although legislation state that mirrors are to be fitted, an issue still persists, that of the mirror has to be checked. With there being six mirrors in total to be checked, leads to a possible situation of an unforeseen object reappearing in a previously checked mirror [42] making the mirrors a solution that lessens the chance of object being missed as of the increased field of view, but does not eliminate the accidents completely. A human factor still remains. One on the side of the driver, and one on the side of the road user.

(48)

Figure 16: Blind Spot Mirrors, picture taken from Ride-along with delivery truck in Skåne, Sweden

Active warning systems that of blind spot detection systems have been standard- ized within many cars, and are available to be fitted within trucks. These systems further reduce accidents. Though the warning, in cases of being fitted in trucks, only resides on the side of the driver, bicyclist as an example still are unaware of the possibly dangerous situation.

4.2 Unstructured Interviews

Unstructured Interviews were held throughout the shadowing sessions during which the topic described in the Method chapter was discussed. Some additional unstructured interviews that were held with the following:

• Project leader for Smart Cities and Regions at Boverket - the National Board of Housing, Building and Planning, Karlskrona, Sweden.

• Staff and administrative personnel at Affärsverken waste management in Karl- skrona, Sweden.

• Head of logistics department of the municipality of Karlskona.

• Staff at Volvo Trucks Repair Center in Arlöv, Sweden.

(49)

4.3. Prototyping 35 What was made clear through the interviews of heavy vehicle operator was that in cases of accidents, it often is not a given that responsibility is split amongst the involved. More so, operators of heavy vehicle suffer the blame. Causes can stem from the fact that the size of the vehicle becomes a determining factor. Operating a larger vehicle, can in the eye of other road users, lead to the mindset that more responsibility lies in maneuvering it in a safer manner. Having the feeling that undeserved guilt can come to be laid upon the heavy vehicle driver, makes for a concerning working situation.

Results of the unstructured interviews also showed that in the case of maneuverability the different stakeholders of city planners and HGV operators each have their own set of needs that do not always match one other. Additionally, with having different legislation, such as dimension factor for roads and planning of housing, as well as the allowed time when HGVs are allowed to operate within cities, all pose difficulties in concluding around a common denominator from which the problem with maneuverability can be solved. Given the time frame of the thesis work, and the resources that were at disposal, led to the decision it not being the ideal field as of the many limitation that were seen to follow.

4.3 Prototyping

Although maneuverability being the most apparent observed problem that heavy vehicle drivers faced, it was was not path chosen to conduct research on. The decision behind this was mainly due to a wide number of factors present within urban cities that come to affect maneuverability in one way or another. Each city is unique in its own way in how streets are shaped. How buildings are planned. Car density as a result of parking spaces available, to name a few. Ultimately, decision making also came down to the limited design space, seemingly having to rethink what defines a truck in order to combat the issue of maneuverability, as urban city structure would be an unchangeable factor.

4.3.1 First Iteration

With this stated, a problem space was set out to be explored. Defined within the frame of visual communication between vehicle and pedestrians, eye contact and driver/vehicle intentions were set out to be improved via clarification through means that could set up heavy vehicles to the autonomous shift, seeing that at some point in time the communication ways between drivers and pedestrians would be to be replaced.

Brainstorming was used in the ideation phase of coming up with possible solutions. An idea of projecting the driver’s sight to the outside of the truck was chosen to be tested. A testing rig was constructed for getting as close to real results as possible, which can be seen in figure 31. The makeshift truck front served as a quick way of adding the experience of a truck front and getting data. Shortening the time for gathering feedback. An early prototype of the eyesight projection was created via having a painted cylindrical plastic rod have strings attached to each end and be freely moving within a plastic transparent water tube, see figure 19. The plastic

(50)

36 Chapter 4. Results rod, representing the position of the truck driver eyesight location, was controlled via having the strings be pulled from the person sitting behind the rig. This setup was tested on a group of students, which all were given a set of questions before given initial impressions and feedback on the concept. Questions, being related to what precautions that are taken before crossing a road, what information is being sought, and if the students have a driver’s license, which could have an impact in what information is being sought and perceived. What was set out to be achieved in asking the student the set of questions, was to see whether the answers stood up to what Sasha presented.

(51)

4.3. Prototyping 37

Figure 17: Truck test rig setup

(52)

Figure 18: First iteration of eyesight projection

Visual Communication Console: Sharing Safety-Critical Information between Heavy Vehicles and Vulnerable Road Users