Proceedings of 8th Transport Research Arena TRA 2020, April 27-30, 2020, Helsinki, Finland
Creating informed public acceptance by a user-centered
human-machine interface for all automated transport modes
Lesley-Ann Mathis
a*, Frederik Diederichs
b, Harald Widlroither
b, Daniele Ruscio
c,
Linda Napoletano
c, Marc René Zofka
d, Alexander Viehl
d, Peter Fröhlich
e,
Johannes Friedrich
f, Anders Lindström
g, Birgitta Thorslund
h, Antonis Litke
i,
Nikolaos Papadakis
i, Nikolaos Bakalos
i, Unai Hernandez-Jayo
j, Stéphane Espié
k,
Viola Cavallo
l, Maria Panou
m, Evangelia Gaitanidou
n, Evangelos Bekiaris
naUniversity of Stuttgart IAT, Nobelstraße 12, 70569 Stuttgart, Germany bFraunhofer IAO, Nobelstraße 12, 70569 Stuttgart, Germany
cDeep Blue, Piazza Buenos Aires 20, 00198 Rome, Italy
dFZI Research Center for Information Technology, Haid-und-Neu-Straße 10-14, 76131 Karlsruhe, Germany eAIT Austrian Institute of Technology, Giefinggasse 2, 1210 Wien, Austria
fTechnical University Berlin, Salzufer 17-19, 10587 Berlin, Germany gVTI, Malvinas väg 6 P.O. Box 55685, 102 15 Stockholm, Sweden
hVTI, Olaus Magnus väg 35, 581 95 Linköping, Sweden iInfili Technologies PC, 62 Kousidi str. Zografou, 15772 Athens, Greece
jUniversity of Deusto - Deusto Institute of Technology, Av Universidades 24, 48007 Bilbao, Spain
kIfsttar/TS2/SIMU&MOTO, Cité Descartes 14-20, Boulevard Newton Champs sur Marne, 77447 Marne la Vallée Cedex 2, France lIfsttar/COSYS/LEPSiS, Cité Descartes 14-20, Boulevard Newton Champs sur Marne, 77447 Marne la Vallée Cedex 2, France
mCentre for Research and Technology Hellas/Hellenic Institute of Transport, Egialias 52, 15121 Athens, Greece
nCentre for Research and Technology Hellas/Hellenic Institute of Transport, 6th Km Charilaou-Thermi Rd., 57001 Thessaloniki, Greece
Abstract
Increasing automation is ongoing in all areas of transport. This raises new challenges for the design and training of Human-Machine Interfaces (HMI) for different user groups. The EU-project Drive2theFuture investigates the needs and wants of transportation users, operators, passengers and passersby to gain their acceptance and to set the ground for a sustainable market introduction of automated transport. This paper describes how HMI concepts for the transport modes road, rail, maritime and aviation in Drive2theFuture are developed and comparatively assessed in order to be able to support an educated use of automated transport. By relying on a stepwise process, adaptable HMI strategies for different user clusters and levels of automation are defined. As a universal method, a comprehensive HMI development toolkit is developed, which can be adopted as training tool to create realistic expectations and enhance acceptance among users, operators and drivers in light of the deployment of automated vehicles.
Keywords: HMI; automated transport; automated vehicles; acceptance; user-centered; training
* Corresponding author. Tel.: +497119702268; E-mail address: lesley-ann.mathis@iat.uni-stuttgart.de
1 1. Introduction
Different types of automated vehicles (AV) gradually enter traffic, implying a fundamental change to the transport system known today. Whereas this opens up many benefits such as a more efficient and reliable urban mobility and a reduction of vehicles on the road, user awareness and acceptance towards automated transport vehicles are a prerequisite for the successful adoption of these emerging technologies (see Hoogendoorn et al. 2014). With regard to automated driving, the EC 2015 Eurobarometer study showed that 61% of potential future users feel uncomfortable with riding in a driverless vehicle (European Commission 2015) and a more recent study in the US showed that 64% of the participants were concerned to share the road with driverless vehicles (Advocates for Highway and Auto Safety 2018). Besides numerous positive experiences with automation, recent accidents also demonstrated how difficult the control of automation can be (Hooper 2013; Yadron and Tynan 2016; Wise 2019). Such accidents gain wide attention. Apart from road traffic, automation also extends to other transportation modes, such as automated rail technology, autonomous shipping as well as unmanned aerial vehicles. Particularly drones have the potential to shape the development of future urban life, given that wide range of potential applications in a smart city exist (Mohammed et al. 2014), but the introduction of a safe and secure automated urban air mobility management has yet to be defined in all its aspects. This shows that many different user groups will be affected by the introduction of automation involving AV operators and riders, railway signalers and traffic management operators as well as non-automated traffic participants. All these user groups share a common means of communication with the automated technology, the Human-Machine Interface (HMI).
As defined by Carsten and Martens (2019) in general terms, the HMI represents a window in human-machine communication, enabling insights into information on mutual intentions and current states. Thus, the HMI plays an essential role for joint operations between users and automated technology. Especially in light of the different automation levels that will co-exist in the future, handover requests from the AV to the human driver and vice versa have to be possible at any time and are subject to intense research (Melcher et al. 2015; Diederichs et al. 2015; Zhang et al. 2019). The human machine interaction with automated systems incorporates special difficulties and challenges that were already very well defined by Bainbridge (1983). She identified ironies of automation and points out that even though automation makes some tasks easier, human attention suffers and new challenges appear when interacting with automated systems. In particular, AVs will have to coexist with conventional vehicles and non-automated traffic participants, such as pedestrians or cyclists. Thus, in-vehicle as well as potential external HMIs have be taken into account when investigating the future interaction with AVs.
The aim of the EU-project Drive2theFuture (“Needs, wants and behaviour of “Drivers” and automated vehicle users today and into the future”) is to determine the needs and expectations of traffic participants facing the deployment of automated transport. Drive2theFuture takes a comprehensive approach by investigating all transport modes (road, rail, maritime and aviation) as well as all types of users (e.g., drivers, travelers, vulnerable road users, fleet operators). In the following, the term “automated vehicle” (AV) therefore refers to cars, powered-two-wheelers (PTW), buses, ships, trains and drones. Since the HMI has the potential to promote a smooth interaction between the user and the machine, one objective of the project is to define the principles of an optimal HMI for the investigated transport modes, levels of automation and clusters of users to enhance acceptance among the public. For this reason, the developed HMI will be tested in different demonstrators in multiple European countries. Moreover, iterative testing and optimization of the HMI concepts with users will be performed. Finally, the identified HMI principles and strategies will be used as input for a software platform, which will provide a method for the assessment of future HMIs in tests and trainings.
2. Challenges in HMI research in different transport modes
There exists very little systematic research on general HMI strategies for AV. In the following, an overview of issues in HMI design in the different transport modes will be given, with a focus on HMI use cases researched in
Drive2theFuture. Since the user’s role will change with the increasing automation of a system, core questions for
all transport modes deal with how users are provided with the appropriate information on the system and how the HMI can ensure situation awareness in case of transitions to manual operation.
2 2.1. Road
In the automotive sector, recent work focuses on HMIs for partial automation, where the human supervises the vehicle (Level 2; SAE 2018) and conditionally automated driving, where the human needs to take over only if the system requests it (Level 3; SAE 2018). HMIs for partial automation in current commercial vehicles mostly use a pictorial steering wheel indicating that automated steering is engaged (Carsten and Martens 2019). As criticized by Carsten and Martens (2019), upon automatic deactivation of the automated steering feature due to system limitations, some HMIs give an auditory warning, whereas others only change the visual icon silently without further notice. This might lead to surprise or confusion of users. Hence, different recommendations and guidelines for the in-vehicle HMI of passenger cars are currently being developed. Recent work by Naujoks et al. (2019) entails an initial set of guidelines focusing on mode indicators for the HMI. These shall ensure safe use and control transitions as well as support users in noticing and understanding these mode indicators. Other recommendations for HMI design have focused on making a Level 3 automated system transparent to the user by displaying appropriate information, for instance by showing when, how and why a maneuver such as a lane change is performed (Debernard et al. 2016). Most importantly, future HMI development involving visual, auditory or haptic alerts will have to find the right balance between conspicuity and annoyance of the information (see Blanco et al. 2015).
Apart from the in-vehicle HMI, previous studies have suggested that new communication needs arising between AVs and other traffic participants, in particular pedestrians, can be addressed by an external HMI (Lundgren et al. 2017; Habibovic et al. 2018). There is a growing research body on awareness and intent communication (Mahadevan et al. 2018; Nguyen et al. 2019), which differ both in presentation modality and complexity (Löcken et al. 2019). Yet, the need for an external HMI as well as its design, displayed information and interaction context are currently not fully understood and still subject to ongoing research (see Moore et al. 2019).
With regard to PTWs, the integration of assistive systems for the rider is not as advanced as in the automotive sector. Previous research has suggested HMIs for advanced rider assistance systems (ARAS) with a focus on haptic cues complementing visual and auditory elements (Bekiaris et al. 2010; Diederichs, Ganzhorn et al. 2010). Haptic HMI elements such as a haptic glove or haptic handle can provide an intuitive and effective way of providing warnings for the rider, leading to high acceptance (Diederichs, Ganzhorn et al. 2010). Even though the large-scale introduction of ARAS for PTW promises to reduce the high number of crashes of PTW riders, the acceptability of systems that interfere with the riding tasks, such as adaptive cruise control or lane keeping assistant, is low among riders (Beanland et al. 2013). Therefore, HMI development for automated functions in PTWs has to put a particular focus on user acceptance. Further questions arising with the introduction of AVs on roads deal with the general low conspicuity of PTWs in traffic, given that motorcycles are vulnerable road users. For instance, additional external lights have shown to improve a PTW’s perceptibility by other car drivers, such assupplemental lights on a PTW’s fork and on the rider’s helmet(Cavallo et al. 2015), which could also enhance detectability of PTWs in future mixed traffic.
2.2. Rail
In rail transport, higher levels of automation have already been realized, e.g., fully automated metro lines without any staff on-board. Nevertheless, a human operator is still required to handle unexpected incidents and to supervise traffic management (Stene 2018). Thus, rail transport automation involves not only train automation itself but also automated traffic management, which monitors rail traffic, identifies problems and reschedules in case of conflict. In Quaglietta et al. (2015), a real-time traffic control system is proposed, able to automatically manage conflicts and deadlocks in case of perturbations. A HMI for the train dispatcher shows the rail network and the planned time-distance trajectories of the running trains as computed by the system, which have to be accepted by the human dispatcher in order to be put in operation. Future research shall focus on the interaction between the automated control system and the dispatcher via the HMI (Quaglietta et al. 2015). Designing user-centered interfaces for dispatchers according to their needs and capabilities is essential for an efficient and safe traffic control and for making sound decisions in case of severe traffic disturbances (Kauppi 2006).
2.3. Maritime
In the maritime sector, autonomous and unmanned ships have been used for decades, for instance in ocean research or military applications (Ahvenjärvi 2016). For unmanned vessels that can be remotely controlled, an HMI is necessary so that an operator can take over control if the automation fails. Several studies have investigated how
3
an HMI for the remote operator could look like. Ahvenjärvi (2002) suggests to include auditory feedback in the remote control HMI, in the sense of a continuous background noise which changes in line with the ship’s operations of rudders and propellers. This way, safety-critical sounds indicating failures pop out more saliently. In Perera et al. (2012), a touch panel is proposed as HMI to control an autonomous surface vessel. It entails different tabs for sensor measurements, GPS display and system controls, including a digital knob for controlling the rudder and propeller position. However, the HMI has only been used in experimental evaluations.
2.4. Aviation
Automation is well established in the aviation sector, being the pioneer of all transport modes. In recent years, the usage of unmanned aerial systems (UAS), usually referred to as drones, has expanded to civilian applications in science and research. They can fly semi-autonomously piloted remotely by a ground operator, but are also able to fly completely autonomously. A study by McCarley and Wickens (2004) revealed that operators caused accidents with drones due to Human Factors issues related to the HMI’s display and control design during take-off and landing phases and in relation to automation system failure. Recent work in the aviation sector suggested also HMI principles for drone operators for fleet management, which has to ensure that the operator can control not only one UAS, but rather several drones simultaneously (Luongo et al. 2019). Therefore, requirements for this kind of HMIs include for instance map visualizations of the aircrafts and their trajectories, mission information, warning signals, display of an artificial horizon and the aircraft’s altitude (Luongo et al. 2019). In addition to graphical user interfaces, a natural user interface for controlling drones was proposed, including speech, hand gestures and body position, which makes multimodal interaction possible in order to reduce visual overload (Fernandez et al. 2016). Future research on the HMI for drone operators will be necessary to ensure safe, efficient and satisfactory interactions for the user.
3. Principles of optimal HMI development
Previous research on HMIs emphasizes the challenges and new requirements arising with an increasing level of transport automation. Drive2theFuture aims at identifying the principles of optimal HMIs for the different transport modes and user groups. The following section describes the stepwise process how the elements of an optimal HMI will be developed. Due to the project’s iterative testing phases, a user-centered evaluation of the HMI principles and concepts will be possible.
3.1. Benchmarking of existing HMIs
In order to take into account existing HMI strategies with regard to automation, Drive2theFuture applies a benchmarking of existing HMI principles from all transport modes: road, rail, aviation and maritime. The goal is to identify good practices in the specific domains and to transfer them to other modes. The benchmarking will address different vehicle types and automation levels. This includes the HMIs from the project’s demonstrator vehicles as well as research and prototypical vehicles of different manufacturers. To perform a benchmarking, defining the attributes to be measured is crucial (Delbridge et al. 1995). We created a questionnaire for data collection (see Fig. 1) that allows a comparison and a clear understanding of the very distinct HMIs alongside a set of attributes independent of the transport modality, including e.g., task description of the automated function, addressed automation levels, feedback on the AV status, available interaction modalities, HMI elements and HMI design. The template shown in Fig. 1 analyzes an example HMI from the EU-project ADAS&ME
(https://www.adasandme.com), describing the HMI function for activation of automation and an emergency
maneuver for a non-reacting driver. The analyzed HMIs will in a second step be expert-rated in order to identify best practices and validate the HMI’s potential user acceptance prior to a first testing phase with users.3.2. Principles for an affective and persuasive HMI
The results of the benchmarking activity will serve as a basis to select and develop HMI concepts for the different transport modalities and automation levels. Drive2theFuture aims at creating affective and persuasive HMIs. Affective states influence processes of perception, cognition and motor control, including attention, situation assessment or decision making (Hudlicka and McNeese 2002) and previous studies have proposed the collection of real time data on affective states to enhance task efficiency and user pleasure during human-machine interaction (Fairclough 2009). Similarly, a user’s attitudes and behavior is influenced by the persuasiveness of the technology
4
(Fogg 2002). An existing theoretical framework for persuasive interface design in the automotive sector (see Paraschivoiu et al. 2019) will be considered in the project’s HMI development process. We therefore regard affection and persuasiveness of the HMI as crucial factors to promote joy of use and acceptance among users. Through iterative assessments on the testing sites and a comparison of the concepts, optimizations regarding the user experience are possible, whereby safety and cost efficiency will also be taken into account. Moreover, the use of standardized key HMI elements such as icons, emoticons, earcons and hybricons for different user clusters, levels and modalities is investigated. Hybricons, a combination of abstract earcon and a representational auditory icon, present a promising approach to convey comprehensible warning signals that are accepted by the user (Diederichs, Marberger, Jordan and Melcher 2010; Diederichs, Marberger and Hinder 2010). Furthermore, emoticons on driver/rider state developed for the HMI framework in the EU-project ADAS&ME are taken into account, which aim at supporting the transitions between automated driving and manual driving (Knauss et al. 2018). By identifying driver states such as happiness, nervousness, sleepiness or anxiety, the system is aware of the user’s capabilities and can adapt to them (Knauss et al. 2018).
3.3. Interaction with other traffic participants
Future traffic will mostly be mixed traffic, in which interactions between AV and other, potentially non-autonomous, traffic participants play a major role. Therefore, this phase will consider HMI concepts addressing interactions with passersby, such as vulnerable road users and cross-modal interactions, e.g., between drones and automated cars in the urban environment. In laboratory and field studies, concepts that engage trust in an AV‘s actions and raise acceptance of non-automated traffic participants shall be identified. In a first phase, operation under baseline conditions without advanced vehicle communication will be investigated, in order to evaluate the actual need for trust formation (compare Moore et al. 2019). In the actual evaluation phase, selected interaction principles will be evaluated on testing sites to collect feedback from involved users.
Fig. 1 Template for the benchmarking with an analysis of a HMI developed in the EU-project ADAS&ME. The definition of the automation levels was adapted from Diederichs (2019)
5 3.4. User assessment during testing phases
In order to draw a comprehensive picture of users’ experiences during the iterative testing phases, we will use objective and subjective measures. A focus is put on using tools giving insights to the emotional responses of users to the AVs during test track and simulator tests.
3.4.1. Objective measurements
Assessing the user experience of autonomous mobility objectively requires the use of sensors to capture a user’s emotional state, e.g., level of anxiety or feeling of fear (Bischoff et al. 2017). To achieve this, wearable devices will be employed that capture targeted biomarkers (blood pressure, heartrate, Galvanic skin response) that are indicators of emotional states. These parameters then need to be correlated with the external stimuli that the user receives during the testing of autonomous solutions both in real and simulation environments, where traffic participants and automated traffic interact safely (Zofka 2018). To achieve this correlation, there is a need to employ techniques that can extract features from multiple modalities, a task to which deep machine learning techniques have been employed successfully in the past (Voulodimos et al.2018). Convolutional Neural Networks (CNN) have exhibited excellent feature extraction capabilities (LeCun et al. 2015), and some adaptations of CNN architectures have shown to be able to classify situations in highly dynamic environments (Bakalos, Voulodimos, Doulamis and Doulamis 2019; Bakalos, Voulodimos, Doulamis, Doulamis, Ostfeld et al. 2019), using multiple data modalities. Moreover, in the simulation scenarios, low-level feature extraction over non-overlapping frame patches and density-based clustering will be used, which are techniques that have been used for real time analysis and classification of video data in previous research (see Papadakis et al. 2019). Overall, these machine learning algorithms classifying the emotional condition will provide insights to the perceived comfort and stress of different user clusters in multiple scenarios.
3.4.2. Subjective measurements
In addition, questionnaires reflecting the user’s subjective experience during the testing phases will complement these objective measurements. One of the subjective measurement tools considered for this purpose is the User Experience Questionnaire (UEQ) by Laugwitz et al. (2008) as it aims at capturing a user’s immediate impression of the interaction with the product and measures user experience in terms of pragmatic qualities, like efficiency and perspicuity, as well as hedonic qualities, like novelty or stimulation. As emphasized previously in Section 3.2, hedonic qualities will play an important role for the project’s HMI development. The questionnaires will be used as an additional modality to further scrutinize the capturing of biomarkers as an indication of user experience. Overall, this setup can provide valuable insights how users of different kind of AV will react in future traffic scenarios.
3.5. Rules for the adaptability and personalization of an HMI
The previously identified HMI concepts and principles will be iteratively tested with different users and in various traffic conditions. Based on this, a cost-efficient process of personalizing and adapting HMI to the needs and expectations of a user cluster in specific conditions will be developed to enhance the acceptance of automated vehicles. For example, the acceptance of advanced driving assistance systems (ADAS), such as forward collision warning or adaptive cruise control, highly depends on a driver’s expectations, skills and preference (Hasenjäger and Wersing 2017). In fact, previous research has shown that personalized ADAS warnings increase drivers’ system acceptance (Panou 2018). Potential factors to be considered within Drive2theFuture include but are not limited to environmental conditions, user cluster, driver and rider state, vehicle type. The outcome of this step will become suggested rules for HMI personalization and adaptation and a conclusion on the circumstances in which these promote enhanced user acceptance.
4. Method for assessing future automation: HMI development toolkit
All principles and strategies identified in the previous steps will be integrated into a comprehensive product, a
HMI development toolkit. This combines the optimized HMI elements, test procedures and user assessment tools
alongside with the adaptability and personalization strategies into a software platform. The complete approach is again summarized in Fig. 2.
6 4.1. Toolkit components and purpose
The software platform will contain the following components: I) a library with optimized HMI elements; II) options for personalizing and adapting the HMI to user needs; III) procedures and user experience assessment tools for HMI testing and evaluation; IV) different use cases and scenarios. The developed software toolkit has a twofold purpose: On the one hand, HMI developers of different transport modalities can use the toolkit to prototype and design a HMI for a chosen use case and target user. On the other hand, the software offers a way to build content for large-scale user acceptance testing and training of HMI concepts for AVs.
4.2. Immersive HMI experience in Virtual Reality
In particular, Virtual Reality (VR) is explored as an environment for the HMI toolkit, giving the users the opportunity to have an immersive experience of encountering and using an AV without being at risk. Previous research on training drivers for take-over-requests in automated driving suggests that VR presents an effective method to prepare drivers for the interaction with an AV (Sportillo et al. 2019). In order to simulate real-life behavior and to contribute to the feeling of immersiveness, the HMI development toolkit will offer interface technologies for haptic recognition and feedback (e.g., steering wheel, joystick).
5. Conclusion and Outlook
The work within the EU-project Drive2theFuture will contribute to raise awareness and enhance informed acceptance of automated transport in the public. The main communication channel between human and automated technology is the HMI. A user-centered, personalized HMI contributing to a user’s affection, persuasion and trustworthiness towards automated transport vehicles can ensure a more sustainable transition phase into mixed traffic with different kinds of AV. In Drive2theFuture, the HMI concepts and principles will be implemented in demonstrators in real world and suitable testing environments for the evaluation with all involved users and transportation modes. Furthermore, a software toolkit for prototyping HMIs for AVs will be developed in the future course of the project, which can provide training options for users of automated transportation systems. For this, emerging VR technologies will be explored to provide users with an immersive experience of future HMI, paving the way for realistic expectations and public acceptance towards AVs.
Acknowledgments
This paper has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No 815001.
References
Advocates for Highway and Auto Safety, 2018. CARAVAN Public Opinion Poll: Driverless Cars. Accessed May 28, 2019 from http://saferoads.org/wp-content/uploads/2018/01/AV-Poll-Report-January-2018-FINAL.pdf
Ahvenjärvi, S., 2002. Audible Feedback for Monitoring of Critical Signals on Ships, Nordic Ergonomics Society ‐ 34th Annual Congress Humans in Complex Environment. Linköping, Sweden.
Fig. 2 Overview of the approach taken in Drive2theFuture to identify the principles for an optimal HMI development of different transport modalities and how these contribute to the HMI development toolkit
7 Ahvenjärvi, S., 2016. The human element and autonomous ships. TransNav: International Journal on Marine Navigation and Safety of Sea
Transportation, 10(3), 517-521.
Bainbridge, L., 1983. Ironies of automation. Automatica, 19(6), 775-779. https://doi.org/10.1016/0005-1098(83)90046-8.
Bakalos, N., Voulodimos, A., Doulamis, A., Doulamis, N., 2019. A Deep-narma Filter for Unusual Behavior Detection from Visual, Thermal and Wireless Signals, ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, United Kingdom, 4005-4009.
Bakalos, N., Voulodimos, A., Doulamis, A., Doulamis, N., Ostfeld, A., Salomons, E., … Li, P., 2019. Protecting Water Infrastructure From Cyber and Physical Threats: Using Multimodal Data Fusion and Adaptive Deep Learning to Monitor Critical Systems. IEEE Signal Processing Magazine 36(2), 36-48.
Beanland, V., Lenné, M. G., Fuessl, E., Oberlader, M., Joshi, S., Bellet, T., ... Yannis, G. 2013. Acceptability of rider assistive systems for powered two-wheelers. Transportation research part F: Traffic psychology and behaviour, 19, 63-76.
Bekiaris, E., Nikolaou, S., Montanari, R., Spadoni, A., Borin, A., Pieve, M., ... Biral, F., 2010. SAFERIDER-advanced telematics for enhancing the safety and comfort of motorcycle riders. In: 17th ITS World Congress on Intelligent Transport Systems, Busan, Korea.
Bischoff, S., Ulrich, C., Dangelmeier, M., Widlroither, H., Diederichs, F., 2017. Emotion Recognition in User-Centered Design for Automotive Interior and Automated Driving. In Proceedings of the Stuttgarter Symposium für Produktentwicklung.
Blanco, M., Atwood, J., Vasquez, H. M., Trimble, T. E., Fitchett, V. L., Radlbeck, J., ... Morgan, J. F., 2015. Human factors evaluation of level 2 and level 3 automated driving concepts: Report No. DOT HS 812 182, Washington, DC: National Highway Traffic Safety Administration.
Carsten, O., Martens, M. H., 2019. How can humans understand their automated cars? HMI principles, problems and solutions. Cognition, Technology & Work, 21(1), 3-20.
Cavallo, V., Ranchet, M., Pinto, M., Espié, S., Vienne, F., Dang, N. T., 2015. Improving car drivers’ perception of motorcycle motion through innovative headlight configurations. Accident Analysis & Prevention, 81, 187-193.
Debernard, S., Chauvin, C., Pokam, R., Langlois, S., 2016. Designing human-machine interface for autonomous vehicles. IFAC-Papers online, 49(19), 609-614.
Delbridge, R., Lowe, J., Oliver, N., 1995. The process of benchmarking: a study from the automotive industry. International Journal of Operations & Production Management, 15(4), 50-62.
Diederichs, F., Marberger, C., Jordan, P., Melcher, V., 2010. Design and assessment of informative auditory warning signals for ADAS. In: Proceedings of FISITA World Automotive Congress 2010, Budapest, Hungary.
Diederichs, F., Marberger, C., Hinder, V., 2010. Iterative design and assessment of an audio visual warning concept for Car2x communication systems. In: Proceedings of the ITS World Congress on Intelligent Transportation Systems.
Diederichs, J. P., Ganzhorn, M., Widlroither, H., Bekiaris, E., Nikolaou, S., Montanari, R., ... Granelli, S., 2010. SAFERIDER HMI Strategies for Motorcycles’ ARAS and OBIS. In 17th ITS World Congress 2010, Busan, Korea.
Diederichs, F., Bischoff, S., Widlroither, H., Reilhac, P., Hottelart, K., Moizard, J., 2015. New HMI concept for an intuitive automated driving experience and enhanced transitions. In: Proceedings from the 7th Conference on Automotive User Interfaces and Interactive Vehicular Interactions, pp. 1-3.
Diederichs, F., 2019, May 22. New levels of automation – now from the user perspective. Presentation at the Autonomous Vehicle Interior Design & Technology Symposium Europe, Stuttgart, Germany.
European Commission, 2015. Autonomous Systems, Special Eurobarometer 427 / Wave EB82.4 – TNS Opinion & Social. Accessed May 28, 2019, from http://ec.europa.eu/commfrontoffice/publicopinion/archives/ebs/ebs_427_en.pdf
Fairclough, S. H., 2009. Fundamentals of physiological computing. Interacting with computers, 21(1-2), 133-145.
Fernandez, R. A. S., Sanchez-Lopez, J. L., Sampedro, C., Bavle, H., Molina, M., Campoy, P., 2016. Natural user interfaces for human-drone multi-modal interaction. In: 2016 International Conference on Unmanned Aircraft Systems (ICUAS) IEEE, pp. 1013-1022.
Fogg, B. J., 2002. Persuasive Technology: Using Computers to Change what We Think and do. Morgan Kaufmann Publishers Inc., San Francisco, CA, USA.
Habibovic, A., Malmsten Lundgren, V., Andersson, J., Klingegård, M., Lagström, T., Sirkka, A, … Larsson P., 2018. Communicating Intent of Automated Vehicles to Pedestrians. Frontiers in Psychology 9, 1336. doi: 10.3389/fpsyg.2018.01336
Hasenjäger, M., Wersing, H., 2017. Personalization in advanced driver assistance systems and autonomous vehicles: A review. In: 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC), pp. 1-7.
Hoogendoorn, R., van Arem, B., Hoogendoom, S., 2014. Automated driving, traffic flow efficiency, and human factors: Literature review. Transportation Research Record, 2422(1), 113-120.
Hooper, J., 2013, July 25. Spanish train crash: automatic braking on notorious curve failed to engage. The Guardian. Accessed May 29, 2019 from https://www.theguardian.com/world/2013/jul/25/santiago-train-derailment-speed-safety
Hudlicka, E., McNeese, M. D., 2002. Assessment of user affective and belief states for interface adaptation: Application to an Air Force pilot task. User Modeling and User-Adapted Interaction, 12(1), 1-47.
Kauppi, A., 2006. A human-computer interaction approach to train traffic control. Doctoral dissertation, Uppsala University.
Knauss, A., Diederichs, F., Wilbrink, M., Lilis, Y., Nikolaou, S., Anund, A., … Finér, S., 2018. An HMI Framework for Driver/Rider States Adaptive Transitions and ADAS. In: Proceedings of the 25th ITS World Congress.
Laugwitz, B., Held, T., Schrepp, M., 2008. Construction and evaluation of a user experience questionnaire. In: Symposium of the Austrian HCI and Usability Engineering Group, Springer, Berlin, Heidelberg, pp. 63-76.
LeCun, Y., Bengio, Y., Hinton, G., 2015. Deep learning. Nature, 521, 7553, pp. 436.
Löcken, A., Gollin, C., Riener, A., 2019. How Should Automated Vehicles Interact with Pedestrians?: A Comparative Analysis of Interaction Concepts in Virtual Reality. In: Proceedings of the 11th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (AutomotiveUI '19). ACM, New York, NY, USA, pp. 262-274. doi: https://doi.org/10.1145/3342197.3344544
Lundgren, V. M., Habibovic, A., Andersson, J., Lagström, T., Nilsson, M., Sirkka, A., ... Saluäär, D., 2017. Will there be new communication needs when introducing automated vehicles to the urban context?. In: Advances in human aspects of transportation. Springer, Cham, pp. 485-497.
Luongo, S., Di Gregorio, M., Vitiello, G., Vozella, A., 2018. Human Machine Interface Issues for Drone Fleet Management. In: International Conference on Human Systems Engineering and Design: Future Trends and Applications. Springer, Cham, pp. 791-796.
8 Mahadevan, K., Somanath, S., Sharlin, E., 2018. Communicating Awareness and Intent in Autonomous Vehicle-Pedestrian Interaction. In:
Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (CHI ’18). ACM, New York, article 429.
McCarley, J.S., Wickens, C.D., 2004. Human Factors Concern in UAV Flight. Institute of Aviation, Aviation Human Factors Division, University of Illinois at Urbana-Champaign.
Melcher, V., Rauh, S., Diederichs, F., Widlroither, H., Bauer, W., 2015. Take-over requests for automated driving. Procedia Manufacturing, 3, 2867-2873.
Mohammed, F., Idries, A., Mohamed, N., Al-Jaroodi, J., Jawhar, I., 2014. UAVs for smart cities: Opportunities and challenges. In: 2014 International Conference on Unmanned Aircraft Systems (ICUAS), Orlando, FL, USA, pp. 267-273.
Moore, D., Currano, R., Strack, G. E., Sirkin, D., 2019. The Case for Implicit External Human-Machine Interfaces for Autonomous Vehicles. In: Proceedings of the 11th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (AutomotiveUI '19). ACM, New York, NY, USA, pp. 295-307. doi: https://doi.org/10.1145/3342197.3345320
Naujoks, F., Wiedemann, K., Schömig, N., Hergeth, S., Keinath, A., 2019. Towards guidelines and verification methods for automated vehicle HMIs. Transportation research part F: Traffic psychology and behaviour, 60, 121-136.
Nguyen, T. T., Holländer, K., Hoggenmueller, M., Parker, C., Tomitsch, M., 2019. Designing for Projection-based Communication between Autonomous Vehicles and Pedestrians. In: Proceedings of the 11th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (AutomotiveUI '19). ACM, New York, NY, USA, pp. 284-294. doi: https://doi.org/10.1145/3342197.3344543 Panou, M. C., 2018. Intelligent personalized ADAS warnings. European Transport Research Review, 10(2), 59.
Papadakis N., Litke A., Doulamis A., Protopapadakis E., Doulamis N., 2019. Multimedia Analysis on User-Generated Content for Safety-Oriented Applications. In: Akhgar B., Bayerl P., Leventakis G. (Eds.) Social Media Strategy in Policing. Security Informatics and Law Enforcement. Springer, Cham.
Paraschivoiu, I., Meschtscherjakov, A., Gärtner, M., Sypniewski, J., 2019. Persuading the Driver: A Framework for Persuasive Interface Design in the Automotive Domain. In: International Conference on Persuasive Technology. Springer, Cham. pp. 128-140.
Perera, L. P., Moreira, L., Santos, F. P., Ferrari, V., Sutulo, S., Soares, C. G., 2012. A navigation and control platform for real-time manoeuvring of autonomous ship models. IFAC Proceedings Volumes, 45(27), 465-470.
Quaglietta, E., Goverde, R. M., Albrecht, T., Jaekel, B., Marliere, G., Pellegrini, P., … Nicholson, G., 2015. Optimal management of railway perturbations by means of an integrated support system for real-time traffic control. In: 6th International conference on Railway Operations Modelling and Analysis, RailTokyo2015, Narashimo, Japan, March 23-26, 2015; Authors version. International Association of Railway Operations Research.
SAE Society of Automotive Engineers International, 2018. J3016: Taxonomy and definitions for terms related to on-road motor vehicle automated driving systems. Warrendale, PA, SAE International.
Sportillo, D., Paljic, A., Ojeda, L., 2019. On-Road Evaluation of Autonomous Driving Training. In: 2019 14th ACM/IEEE International Conference on Human-Robot Interaction (HRI), pp. 182-190.
Stene, T. M., 2018. Automation of the Rail – Removing the Human Factor?, in “Safety and Reliability - Safe Societies in a Changing World”. Proceedings of ESREL 2018, Trondheim, Norway. In: Haugen et al. (Eds.). Taylor & Francis, London, pp.1947.
Voulodimos, A., Doulamis, N., Doulamis, A., Protopapadakis, E., 2018. Deep learning for computer vision: A brief review. Computational intelligence and neuroscience. vol. 2018, article ID 7068349.
Wise, J., 2019, March 14. The Boeing 737 Max and the Problems Autopilot Can’t Solve. The New York Times. Accessed May 29, 2019 from https://www.nytimes.com/2019/03/14/opinion/business-economics/boeing-737-max.html
Yadron, D., Tynan, D., 2016, July 1. Tesla driver dies in first fatal crash while using autopilot mode. The Guardian. Accessed May 29, 2019 from https://www.theguardian.com/technology/2016/jun/30/tesla-autopilot-death-self-driving-car-elon-musk
Zhang, B., de Winter, J., Varotto, S. F., Happee, R., Martens, M., 2019. Determinants of take-over time from automated driving: A meta-analysis of 129 studies. Transportation Research Part F: Traffic Psychology and Behaviour, 64. doi:10.1016/j.trf.2019.04.020.
Zofka, M.R., Ulbrich S., Karl, D., Fleck T., Kohlhaas, R., Roennau, A., Dillmann, R., 2018. Traffic Participants in the Loop: A Mixed Reality-Based Interaction Testbed for the Verification and Validation of Autonomous Vehicles, 2018 IEEE International Conference on Intelligent Transportation Systems (ITSC), Maui, HI. 3583-3590.
