Radical Innovation and Design for Connected and Automated Vehicles

Moderated by Prof. Riener, a panel consisting of Regina Bernhaupt, Mark Colley, Andreas Löcken, and Nikolas Martelaro discussed the possibilities of future users of automated mobility. In this panel, questions regarding current challenges and possible avenues to a more equitable future were discussed. The questions were especially:

• $\mathrm{■}$

  What new user groups can automated driving open up?

• $\mathrm{■}$

  What would be prerequisites for a successful integration into future transport?

• $\mathrm{■}$

  What accessibility problems are currently present in private/public transport?

• $\mathrm{■}$

  What would be a radical innovation for blind people in future mobility?

• $\mathrm{■}$

  How to ensure all people have access to future mobility?

The panel and the attendees identified interesting avenues for future methodical, educational, and human-factors-related research areas. Especially relevant are the requisites of designing not for a “standard” user but for a user range. The according tools and evaluation protocols for such designs were proposed as a highly relevant research direction. This includes software-defined vehicles both for appropriating hard- as well as software needs.

Ulm University with Human-Computer Interaction Institute – Carnegie Mellon University – Nikolas Martelaro
Based on a common research interest in supporting the analysis of driving-related data, the future development of an analysis tool for rich data and metadata in the automotive domain is envisioned. This collaboration could include requirements analysis of the tool, workshops on usability enhancement, and data sharing for the advancement of understanding user behavior both in manual and automated driving.

Ulm University with Centre for Accident Research & Road Safety, Queensland University of Technology, Brisbane, CRICOS 00213J, Australia – Ronald Schroeter
The Centre for Accident Research & Road Safety currently evaluates automated vehicles and their behavioral impact. This includes the recording of numerous variables related to the driving task as well as the driver or user state. Prior to this, a vehicle was developed that is allowed to drive on various roads. The rich information logged by these vehicles can be used for deeper analysis and to guide future research in domains such as computer vision or interaction with automated vehicles.

Ulm University with RISE Research Institutes of Sweden – Jonas Andersson
Until automation capabilities become so advanced that a truly autonomous service is supported, that is, path planning, path execution, interaction with vehicle users, moderation for varying needs, handling of all imaginable situations, and maintenance work planning, some degree of human supervision is necessary. Currently, the arrival of this future is not predictable. Therefore, to still benefit of the advancements in technology and with the pitfalls of supervising automation in mind, remote control is developed and evaluated both from a technical and a Human Factors perspective. In this potential collaboration, remote control interfaces intended for multi-vehicle and/or multi-supervisor should be developed, implemented, and evaluated.

Ulm University with Cornell Tech – Wendy Ju and David Goedicke
Based on the simulators developed by David Goedicke and Wendy Ju at Cornell Tech, the evaluation of numerous aspects regarding the interaction with an (automated) vehicle is possible. For example, a real vehicle equipped with AR glasses can be used to study braking behavior for simulated crossings of pedestrians, the interaction with real forces within the vehicle, or as a tool to use with a Wizard-of-Oz approach for the participants to take the place of the pedestrian. This enables the evaluation and assessment of numerous concepts designed and evaluated in Virtual Reality.

When attending a workshop named Radical Innovation and Design for Connected and Automated Vehicles one of the key expectations is to find ways on how to design for connected an automated vehicles that will very likely be available in a not so distant future. What was impressive in our workshop was the change towards even more critically reflecting on the assumptions we make for such a radical shift. Where would be the revolution – and not only the small evolution.

Reflecting on develpments in socio-technical systems and on their design in the past 30 years, we structured our thoughts on three dimensions: (1) users and groups of users, (2) systems and technology, and (3) the stakeholders or companies. We were able to identify a set of critical aspects and assumptions that need to be discussed and focused on in the near future: The need of negotiation between stakeholders and companies and the users, in terms of overall goals for transportation and mobility, be it monetary or based on other key characteristics of society like health, well-being or sustainability.

For the planning horizon of 15 – 20 years it became clear to me, that this lovely and inspiring group in this workshop is currently designing technology that will be key in my early retirement years. This led to a shift in my reflections and focus: while a 15 to 20 year horizon is key for a researcher mid-career, my research contribution should be more on the evolution and the more immediate improvement as well as the lay-out of a framework that can guide today’s discussion.

Based on a number of brainstorming sessions and group work, I came up with a triangle depicting user, system and stakeholders and allowing to group the various discussion points and possible conflicting requirements in one central figure. It can be a basis to summarize the contributions and might lead to a better communication for researchers with varying research backgrounds (as experienced in the Dagstuhl seminar).

Regina Bernhaupt
Professor for Digital-Physical Data Interaction (Systemic Change)
Industrial Design
Eindhoven University of Technology

It is interesting to observe how the core focus of our discussion is gradually shifting from designing for interactions and interfaces of the personal automobile to designing for a holistic end-to-end experience of smart and sustainable mobility. The future of mobility points increasingly in the direction of shared mobility and the gradual phasing out of the personal car. Public transport offer several benefits such as sustainability, better economic viability, and reduced environmental impact. However, this also comes at a cost of aspects such as convenience, flexibility, privacy, and speed. Here, the question arises: to what extent can the “sacrifices” to personal convenience and comfort be addressed and/or mitigated?   The benefits of end-to-end, seamless mobility afforded by personal vehicles soon disappear when adapting to the currently fragmented nature of public transport that is often plagued by unpredictability, lack of flexibility, and lack of support for first/last mile conveyance. Ignoring the emotional aspects of car ownership (in terms of attachment to a car, or the thrill and feeling of freedom), I believe that these core practicality aspects of individual car ownership needs to be addressed for us to be even remotely successful in realizing this vision. I see this therefore as an interesting but uphill challenge (problem statement) for us interaction designers, HCI researcher, and policy experts to consider and address. Undoubtedly, the next decade(s) will see some unprecedented innovations in terms of mobility (combination of new powertrains/electric propulsion, connected and automated driving, and MaaS) leading to radical disruptions in the realization of sustainable mobility.

The activities in the workshop focused on understanding how transportation can be solved in this new environment. A new environment that requires is filled with companies pushing for automation, a public that demands “greener” technologies, and a global political situation that restricts access to traditional energy sources.

Through discussion with the different experts that joined the conference, it became apparent that a new radically new perspective is necessary. The future of individual transport was taken under scrutiny and modern concepts of transport were explored and envisioned (like on-call bus services, ride-sharing concepts etc.).

Even though the perspective was on finding modern solutions, a substantial amount of time was also spent on disassembling techno-optimist solutions to the current transportation problem. It became evident that car manufacturers are having difficulty changing their strategy. Modern idealized transportation concepts are often present in a simplified view and, even worse, often ignore marginalized uses of private transport (Home health workers).

The activities in the seminar, from walks to movie nights to board games, created an overall collegial environment and even a playing field where everyone contributed. The fact that multiple videos produced during this week are planned to be submitted to the AutoUI conference speaks to the progress made this week.

In this work we show novel approaches to fully automated driving.

On the mechanical side, we propose to build modular cars which are split in multiple subsystems and allow for a multitude of different combinations to fit the user, usage context, application area and environment. In addition, we present the idea of a digital agent which helps travelers to decide on means of transportation in accordance to their individual preferences, such as using green modes of transport, their budget, supporting privacy during the journey or travelling as fast or convenient as possible. The focus of the agent lies on seamless interaction of different modes of transport where cars driven on fossil fuels get excluded as much as possible.

We believe these ideas form a base for future research in the mobility domain.

The key insight I gained from the Dagstuhl workshop on Radical Innovation in Connected and Autonomous Vehicles is that we need to re-invent the experience people have with transportation. Currently, people are solving a lot of their problems with the “silver bullet” of their personally owned car. They might appreciate that this solution comes at some cost – to the environment, to their health, to their everyday routines – but also it allows them to get to work, to explore the world, to meet with people, and to enjoy their lives.

I believe it is possible to help transition people to finding new ways to solve their problems, so that they might reduce or remove their vehicle usage, to the point that they might not need personally owned vehicles at all. To do this, though, people need information about the myriad options for transportation and confidence about their ability to adopt new modes of mobility. To this end, I believe that we should consider ways to improve self-efficacy in mobility. Perhaps, for example, people need a travel concierge to help them brainstorm new ways to travel, and help them through the “but but but”s that keep them from trying and then adopting new travel. Key aspects of this need is to provide people with specific solutions to solve their specific issues – such as needing to transport children, or needing to arrive quickly or cheaply, being able to fit a meal or a workout.

Over the course of the workshop, we learned many ways of exploring and expanding ideas – design thinking, user enactments, remote ethnography, video prototyping, collecting field data. These skills were put to use in the development of an idea for a virtual agent to help people plan travel without personal cars. The super group of Natasha Merat, David Sirkin, rOnny Schroeter, Lutz Morich, Virpi Roto, Kai Hollander, and myself made a humorous video prototype of a virtual agent that aids future mobility for a person with disabilities who is planning a bank heist. I am excited to see the way that this group will build on the skills and methods learned in this week to tackle the great challenges that lie ahead in transforming people’s experience of mobility.

Current challenges such as climate or demographic change point to a radically different future. To maintain (or even improve) our standard of living, we need to keep up. And what could be more radical (for automotive human-computer interaction researchers) than removing cars from traffic? Automated shuttle buses or personal “transportation-pods” may replace personally owned vehicles while being more sustainable and enabling more people to participate in traffic. In our own research, we focused a lot on how the automated vehicle should communicate with its passengers (e.g., [1]) or its surroundings (e.g., [2]. But will people accept “giving up” their cars? In my opinion, the most interesting challenges for a radically new future of mobility (that HCI researchers can tackle) will be to create inclusive solutions (and not “just” add accessibility options) and meet user experience needs that manually controlled vehicles will no longer provide. If we cannot give users an alternative to their cars, the industry has no incentive to provide the technology or stakeholders to update policies. The answer to the title is: “probably not”. I look forward to continuing the discussion in future workshops and seminars. I thank Wendy, Bastian, Andreas and the Dagstuhl organization for allowing me to make new contacts and deepen existing connections.

To help reduce the impact of automobiles on traffic congestion and the environment, I believe there needs to be a shift towards smaller vehicles. To make this more agreeable to people, I also believe that there needs to be ways to make sure that such vehicles meet people’s needs. I suggest that this could be done through new design systems that help people to radically personalize their vehicles for their needs.

A current issue with today’s automobiles is that they are physically too large. As people begin to shift towards electric vehicles, I believe that we are also starting to further understand what our driving needs actually are and are thus changing what we believe we need in a vehicle. For example, a primary issue that many people have had with electric vehicles is they have too little range for their driving behavior. While this is true for some drivers, it is not true for many drivers, especially those who live in cities, as they may only drive short distances and thus would be ok with smaller vehicles with less range. Moreover, with the convenience of charging at home (for some), range becomes a low priority issue once a certain threshold of range capacity is reached (I hypothesize that this is about 240 KM of range). While new battery technologies that allow more dense energy storage should be explored, I believe that there is not as much of a need for creating physically larger vehicles to meet consumers; transportation needs.

If we can reduce the size of vehicles then, we can also reduce their impact on congestion on the road and while parked. While smaller vehicles are much more common in Europe, in the US, vehicle design is trending larger and larger. Much of this is likely based on consumer psychology as people get a vehicle for the 5% of driving that they do rather than the 95% of driving. For so long, vehicles were affordable enough that people could own a vehicle to express their identity. However, with rising manufacturing costs, vehicles are now going to be more expensive, making it harder for people to own the vehicles that they may desire. One way to combat such rising prices though is to reduce the material costs of a vehicle, for example by making smaller vehicles. Still, such small vehicles will need to have great designers in order to fulfill people’s physical needs as well as their psychological and aspirational needs. I believe that one radical further for personal automobiles is in highly personalized, ultra-compact cars that are designed specifically for every person to fit their needs. Such a future would likely require a few key technical innovations that require research and development.

First, EV drivetrains will likely need to be modularized so that the most expensive parts of a vehicle can be mass-produced and interchangeable with different body styles. Such technology is already in development with the concept of EV drivetrain “skateboard” platforms that allow different body styles to be bolted onto the skateboard frame. Second, we will need better ways of determining people’s physical needs in regards to range and functionality of a vehicle. While one may assume that people determine their own physical needs when choosing a vehicle, I am not as sure about this and believe that many people consider such needs but in more aspirational or speculative ways. Often, people drive alone on daily commutes and often do not take the long road trip with many people or haul lots of gear to go camping. I believe that there is research to do on how well people vehicle choices meet their physical needs. Such research might include quantified self research around driving habits and ethnographic studies at the scale of people’s use of their cars. I hypothesize that many cars are highly underutilized. Based on such research, I also believe that such ethnographic and data collection systems can be used for the design of new kinds of vehicles that do fit people needs. To accomplish such personalized design on a massive scale would require new kinds of design tools and likely computational assistance. Research into generative design systems may lead to new tools that can allow a person paired with a computational “designer” to create a car body/interior design just for them. Such systems could design vehicles for people who do transport many other people, or who go on trips to the woods to ski and camp, or who want to have a stylish vehicle to show off their personal tastes. One can even imagine that as people’s needs change over time they can change out their bolt-on bodies for new ones while keeping the same skateboard frame.

Such an idea is not without its potential issues and here I try to describe some that may need to be overcome when considering such a personalized automotive design future. First, what happens when a vehicle does not meet someone’s needs any more and must be resold? Will it be so personalized that it does not work for another user? Or will there need to be a service to find similar enough users who could then purchase a customized vehicle from another person? Or, are there ways to make modifications on a customized vehicle that allow someone to keep most of what is there and change out only key areas? Another issue is that of fulfilling driver’s needs when they do take the 5% of trips where their vehicle is not a good fit? This might be overcome with some innovation in car sharing / rental business models where people can rent the car they need for the few times a year that they need it. Finally, there is a question as to if more investment should be made into automobile travel or into other forms of transportation. I personally believe that we will need a diverse set of transportation options well into the future, however, such technologies that can help to understand people’s needs and personalize their car might also be the foundation for systems that can design personalized mobility plans for people that includes personal vehicles, public transportation, bicycling, and other forms of micromobility.

When debating future mobility concepts there is always a tension between the doable and thinkable. One way how to look at the future is to expand it from the past and think of how the future may evolve. In the context of automated vehicles this is often done through the lens of automation levels. Coming from manually driven vehicles (level 0) lateral and longitudinal control of the vehicle, as well as the necessity to observe and react to the environment are omitted. At level 1 the vehicles accelerate and break by means of ACCs or keeps the lane by utilizing lane-keeping assistance systems. On level higher the vehicle does the driving and the driver “only” needs to observe the environment and react if necessary. On level 3 non-driving related tasks (NDTR) are possible, but the driver needs to react to take-over-requests (TORs). Higher level (i.e. 4 and 5) even relives drivers from any driving activity allowing them (maybe) to intervene in tactical driving decisions (e.g. shall the vehicle overtake or not) and strategic ones (e.g. telling the automated vehicle where to go). Increasing levels allow us to think of different kinds of transport systems. Vehicles evolve and become more and more autonomous. This is one way to do it.

Another way is to come from another point of view: thinking of and exploring solutions by asking “What if” questions. This allows us to re-think the future of mobility more radically. But what are good “What if” questions? How far do we have to go to be radical enough? Whom do we have to involve in this process? This Dagstuhl seminar was a starting point for me to think about radical changes in our premises leaving (hopefully) to novel insights on how the future of mobility could look like. Questions that have been (or will be) tackled are the following:

• $\mathrm{■}$

  What if all public transport would be free of costs for people who use it?

• $\mathrm{■}$

  What if we prohibit people of owning their own vehicle?

• $\mathrm{■}$

  What if we forbid individual cars at all?

• $\mathrm{■}$

  What if we get rid of all streets?

• $\mathrm{■}$

  What if we forbid manually driven vehicles?

By making this “what if” statements to our premise we were forced to come up with radical new solutions but also by looking at problems that may arise for individuals triggered by these radical changes.

As a novice and industry participant at a Dagstuhl seminar, all estimates and impressions are based on an absolute jumping off basis without relative comparison to comparable seminars.
My goals before the seminar were the following:

1. 1.

   broaden my perspective regarding mobility (narrow due to industrial background)

2. 2.

   networking with mobility research experts in academia

3. 3.

   learning methods for innovation and knowledge generation

My main takeaways: My expectations were exceeded by the very diverse group from the science of HCI in combination with mobility issues. My dull glasses of industrial imprinting have wich not only cleared up but also significantly expanded. Through the dynamic composition of working groups in connection with challenging and innovative methods, both my goal of networking and methodological competence enhancement was fulfilled. Connections and perspectives have emerged that promise a closer cooperation of my private economic background with science in the field of automated and networked mobility in the future in a general and concrete way.

Lutz Morich
 frE3-Innovations
Josef-Fleischmann-Str. 7
85139 Wettstetten
Tel.: +491605739144
Lutz.Morich@fre3-innovations.com

Human mobility has massively changed throughout the last centuries. Humans had used horses (or similar animals), horse-drawn carriages, or just their own feet to get to a destination for thousands of years. This hugely limited individual mobility and only with the invention of bicycles, trains, cars, busses, and trucks humans gained mobile flexibility and this massively extended their motion range. Vehicle automation, climate change, and novel drive technology are expected to again change to a large extent how we will use and experience mobility.

There is a variety of concepts for automated cars and how mobility could look like in the future (e.g., 2030 or 2050). This includes concepts of automated individual vehicles and shared mobility using various types of vehicles. What many of these concepts have in common with current research and development practices is that they rely on rather incremental and evolutionary steps. An open question is whether this approach is suitable to use or whether radical innovation and design could be more suitable.

Looking at other fields such as communication industry and especially phones, the history of the phone shows that a similar evolutionary approach was used multiple times: operators were replaced by machine switching, wireless phones at home replaced traditional fixed phones (=the user got rid of the phone cord) and later traditional mobile phones (“feature phones”) offered mobility to their users as they were able to use their phone not only at a fixed location but everywhere. For quite a while, such phones were improved incrementally to satisfy specific user needs (e.g., longer standby time, better sound quality, …). The introduction of the smartphone instead can be seen as a more revolutionary step, since prior achievements or principles (e.g., standby time, privacy) were sacrificed for a novel type of device (large touch screen) and an ever-increasing App ecosystem. As history now tells, the gained benefits of smartphones outweighed the user need for week-long standby times and former global players, who sticked mostly to their good old feature phones, became mostly meaningless with regard to phone production. Another interesting aspect in communication industry can be found when looking at today’s business models for mobile phone (operating) systems when looking at the two biggest players and ecosystems: What both have in common is that they offer large App ecosystem. One of the biggest difference between them is that one generates revenue from selling a combination of hardware (phone) and operating system, while the other one mostly offers a free operating system (leaving production of hardware to other companies) while generating revenue from collecting and using data with all its pros and cons.

The previous case of the mobile phone sector, where the business has change massively during the last decades, raises a few questions for the automotive field:

• $\mathrm{■}$

  What does it need to prevent current players in (automotive) mobility from becoming the “next Nokia” and instead contribute successfully to novel forms of mobility?

• $\mathrm{■}$

  What are the consequences of different business models and how do they affect society?

• $\mathrm{■}$

  How can we ensure to develop future mobility in an ecological way where users, society, and environment are first and business/revenue are only second?

Taking the experience from the smartphone sector into account, also related to market power, I consider it essential that we do not make the same mistakes again that happened in the phone industry when it comes to (re-)inventing future mobility. Thus, I hope that the discussion and ideation of how our mobility will look like in the next 10, 30, or 50 years happens equally in industry, society, and academia. With the outcomes of this Dagstuhl seminar, I hope that we can contribute to and stimulate this discussion.

I thoroughly enjoyed to be back at Dagstuhl in person and discuss our ideas and perspectives with old and new collaborators. What I found remarkable is that the focus of the seminar shifted very quickly into novel directions such as “what if … the individual car will/needs to disappear” and it was a pleasure to discuss creative solutions and approaches to radical changes in the mobility sector with all attendees.

We met for a week in Dagstuhl to discuss radical innovation approaches to change future transport, to identify problems, and to develop solutions. The aspects and dimensions brought in were diverse and the discussions from different perspectives very fruitful.

But we also realized how difficult it is to break out of the “normal” iterative development process and think “radically” or “innovative”; after numerous presentations, discussion rounds, panels, prototyping workshops and, of course, evenings in the wine cellar of Schloß Dagstuhl, we gave birth to great ideas to revolutionize (individual) transport in the future. At the same time, however, we have also realized that there are a number of obstacles to be cleared out on the way to more sustainable mobility. On the one hand, car manufacturers and the automotive lobby are strong advocates of individual cars and it will be difficult to initiate the rethinking process there. On the other hand, political decision-makers also need to be convinced to launch new forms of mobility.

Today, however, at the end of the seminar, we are confident that we will be able to influence future decisions and implement some of the ideas developed in the seminar. More on the short-term, some ideas for bilateral/multilateral collaborations, research exchanges and last but not least, conference papers and grant proposal have emerged.

Thank you for the opportunity to organize this seminar in Dagstuhl and to let all these creative, innovative minds work on problems of the future for one week and to foster personal interaction after a long time of remote meetings.

The Game is NOT Over – we are coming back!

Andreas Riener Co-organizer of Dagstuhl Seminar 22222

Radical innovation requires powerful innovation methods. Through some years of development and testing, we have developed the concept of experience goal and the method of experience goal -driven design. In this seminar, we experimented using the method on innovating future vehicles. Pairs of participants started by defining experience goals for a given use case, e.g., Eric, 75y, having Covid-31 and needing transportation to hospital. What kind of emotional needs does he have? An experience goal should communicate the experience or psychological need that the person has in the situation. The experience goal should then drive the idea generation: what kind of future vehicle could enable the experience goal to realize? In this case, the group found it is important to support feelings of safety and being nurtured. The social needs include contacting the closest people and informing about the status.

Mobility is both a desire and a necessity. Mobility practices, such as commuting to work, shopping, or, traveling to visit loved ones are an integral part of our everyday lives. Currently, a large proportion of these practices are performed through manually driven vehicles. With the rise of driving automation, however, these practices will change. Driving automation promises to relieve people from driving-related tasks and provide them with the opportunity to engage in non-driving-related activities. To fulfill this, of course, the automated vehicles should ensure safety and reliable maneuvering which requires systems that are able to sense and understand the environment and act flawlessly. This pragmatic aspect of automated driving has been researched in the last decades. It is nevertheless, important to investigate how these automated vehicles change the mobility practices and what experiences emerge through their use. This as well raises the question of whether the new practices are accepted by users and can be integrated into their everyday lives. While the decisions related to the design of technology for automated vehicles are mostly technical, they have social consequences. Creating technology that forces users to follow rules that the designers have planned for them without considering their individual needs or social norms can create a host of negative experiences. Therefore, the design of automated vehicles should consider the individual and social experiential aspects besides the pragmatic and safety-oriented factors.

My interests are to make Automated Vehicles safe and fun to use, so I attended the seminar to be inspired to explore new ideas to that end. Here is what I walked away with: Let’s apply a “dog” metaphor to future Automated Vehicle (AV) design! Dogs have an innate connection to human behavior. They are receptive to their owner’s cues, sense their owner’s feelings or moods (or “states”), adjust or tailor their behaviour accordingly, and even probe their owners for explicit cues or responses when needed (by lifting paws, making sounds, nudging, etc.). As a result, humans develop a deep connection to their dog. How can we create such deep connections in the context of human-AV interaction, and opportunities for mutual learning processes?

We propose a vehicle decision-making framework that incorporates cues obtained from the vehicle user and an adaptive multimodal HMI (human-machine-interface, considering all human senses, so including vehicle dynamics) that probes the user. The framework integrates the user’s state (attention, arousal, stress, …) and the external environment (complexity, planned path & maneuvers) while making operational decisions, and applies active interventions on user states. This bi-directional, adaptive tuning mechanism or “language” between AV and user, may then potentially be used to

• $\mathrm{■}$

  enhance safety (aligning cognitive state with environment requirements),

• $\mathrm{■}$

  improve user experience (aligning with or playfully manipulating emotional states),

• $\mathrm{■}$

  create user profiles that optimise AV behaviour, and

• $\mathrm{■}$

  improve user experience.

Fully or partially autonomous vehicles are a rich but vexing context for interaction designers to study in real road contexts. On one hand, their design draws on established lines of research in physicality and human factors, and in psychology and perception. On the other hand, for now, real autonomous vehicles are difficult to acquire and program, and thus apply to interaction design research questions. One approach is to use Wizard of Oz prototyping, typically hiding a human driver, which can work for participants both inside and outside of the vehicle.

Monday’s tutorial focused on collecting field data in the context of Wizard of Oz prototyping of human-vehicle interactions in the field. We covered several examples of prior studies, which led to both the kinds of findings we were looking for1 (about deceleration profiles, pedestrian perceptions) and those we had not expected2 (that some road users “grief” autonomous vehicles). These experiences suggested the value of many hours spent selecting and driving in various contexts to capture a breadth of behaviors. The conversation quickly moved to sharing practical tips during field work, such as the amount of time needed to plan a field study, the daily setup and breakdown of equipment, a suitable number of interviewers for each outing, alternative ways to design questionnaires, and how to increase participation from passers-by, such as offering to walk along with those who are reluctant to stop and talk with interviewers.

My personal goals for the program were to meet human factors experts, computer scientists, engineers, and other designers, share our research methods, and plan future collaborations. While the program had many organized activities, these 2 goals were those where I focused most of my effort and have already found the most value.

My research interest is driven by a vision of sustainable everyday mobility, where mobility is something that allows people to go where they need to be and do what they need to do, and where people have equal access to functions in society, healthy environment, and safe, appropriate, and effective ways of getting around. Partial or full automation of vehicles can play a part in facilitating such a future vision, but it is necessary to investigate how and where to best implement automation so that society can gain the most benefits from it – to potentially radically rethink the way the mobility system is built today.

Currently, there is a strong technology push in relation to automation, and a weak societal pull or engagement. There are risks associated with this dynamic, as technology becomes implemented where it is possible instead of where it is beneficial. Research and industry efforts are then spent trying to handle the resulting consequences of this sub-optimal implementation, including overtrust in automation, managing sudden and dangerous handovers and the difficulties in communicating among different road users. There is a need to take a step back and look at the interplay between humans and automated systems from a more comprehensive and multi-stakeholder perspective. Which role(s) in mobility should automation play, and which should be played by humans? During use, which actions should be done by whom? And who should make which decisions and take which responsibilities? Taking a step back will open up the design space to not only create radical technical inventions, but also truly radical innovations that are implemented in a way that benefits both society and the environment.

A radical rethink requires a combination of critical and creative approaches, and multiple perspectives. My motivation to join the seminar is to meet new people (and old acquaintances) to get new ideas for approaches and clash my perspectives against others, and in the long term create new collaborations. I am especially interested methods that allow us to explore an unknown future in such a way that we can explore the consequences for interaction, use and society at large. My previous experiences include enacted and real-world trials of mobility solutions, with the intended users in their intended contexts, as well as more future-oriented and speculative approaches using provocative enactments and prototypes to evoke a discussion on the desired future. It would be interesting to find other ways of exploring future practices, to predict, enact and evaluate that practice in advance, in order to understand if ideas will be useful and valuable in the future.

This seminar has explored multiple ways of doing so. At the same time, the activities and mix of interests represented has pushed us into a simultaneously critical and open mind to the visions of the future available in the world or that we can come up with ourselves. This fruitful exercise has allowed us to find the faults in our thinking, challenge our assumptions and realise who we have included and excluded in the future, Importantly, this leads to the realisation that people will still be people in the future, lives will be messy and any radical mobility solution must cater to many conflicting interests

I come away from the seminar with a renewed interest for research into the experiences and relationships that forms with, within and beyond the vehicle, validating the work I and my research group have investigated previously. I am also empowered by the discussions that more people (in a car dominated research area) feel the need to challenge the car or even get rid of it altogether. However, it is important not to lose track of the important role that personal cars can play – it is, and most likely will be, a part of the diversity of solutions that are a key to successful sustainable everyday mobility in the future.

We further wanted to use the seminar as an opportunity to communicate new design and usability evaluation methods. This activity was also organized as a community event – seminar participants could suggest methods that they find particularly innovative and profitable and that they would be willing to communicate to interested parties in a short session (15 min. to 90 min.) prior to arrival. Feedback was clustered and finally offered in 8 small groups on Monday afternoon and Tuesday morning (each seminar participant could participate in several activities). Here is an overview of idea generators and topics.

• $\mathrm{■}$

  Teach Design Methods, part I (Monday, 14.00-15.30 hrs)

  1. 1.

     Virpi Roto: Experience goal-driven design (90 min.)

  2. 2.

     David Sirkin: Collecting field data (from passers-by) (90 min.)

  3. 3.

     Gary Burnett: VR to run design workshops (90 min.)

• $\mathrm{■}$

  Teach Design Methods, part II (Tuesday, 9.00-10.30 hrs)

  1. 1.

     Nikolas Martelaro: Remote Wizard of Oz methods (for automotive interaction) (60 min.)

  2. 2.

     Andreas Riener: User Enactments as a Design Method (60 min.)

  3. 3.

     Regina Bernhaupt: Design-thinking workshop (60 min.)

  4. 4.

     from 10.00AM (for everybody) David Goedicke: Driving Simulator Demo

  5. 5.

     from 10.15AM Ronald Schroeter and Nik Martelero: What could go wrong? An interactive card game (two groups)

For the World Cafe event, we had people brainstorm topics they wanted to talk about with the entire group, and then grouped these. From the groupings, three main topics were selected.

Nik Martelaro, Shadan Sadeghian Borojeni and Regina Bernhaupt lead each of these themes. The test of the team broke into three groups, which rotated through each theme discussion. From this, we created a shared group sentiment on the area, interesting research questions and desired outcomes.

Nikolas Martelaro’s theme focused on Personalized Design for Future Mobility. The advent of new forms of vehicles which could have modular blocks would make it possible to adapt for users who had different abilities, people who are on extremes of the human factor sizes, and families, who might have dogs or children. The modularity would make it possible to have reconfiguration for long trips or off-roading. Key questions are who is legally liable for manufacturing and safety validation for each vehicle?

Shadan led the discussion for new models and paradigms for interacting with automation. The research questions stemming from this area were:

• $\mathrm{■}$

  RQ1: what are the needs for new models and paradigms? Potential answers were, new control schemes, new relations, expectation management/mental models, and responsibility management.

• $\mathrm{■}$

  RQ2: what are the implications of different models and paradigms for identity and for breakdowns in interaction.

• $\mathrm{■}$

  RQ3: How do we convey roles and responsibilities in changing levels of automation?

• $\mathrm{■}$

  RQ4: How acceptable are the defined roles for different users?

• $\mathrm{■}$

  RQ5: How to keep up with fast updates of car?

• $\mathrm{■}$

  RQ6: How do different interaction paradigms shape emerging mobility experiences?

Regina led a discussion of reinventing the automotive experience. The key stakeholders in this were the government, users and technology providers. The government is responsible for thinking about the economic impacts of mobility, the public health implications, and sustainability. The technology providers are thinking about the physical, emergent and virtual aspects of mobility technology. Users, on the other hand, are thinking about self-efficacy, user experience and their personal values. Interaction between technology and users is largely around automation and mobility services. Interaction between the government and technology companies have to do with policy and standards. And the government interacts with users through training and licensure. Key ideas in the arena have to do with improving self-efficacy for individual users, groups and community, having virtual butlers who can help individuals to customize their transportation, and visions of post-car transportation from the technology providers.