UI/UX for Autonomous Vehicle interface to build trust.
The UI/UX was created as a part of a Human-AV Connect Design Challenge. The challenge was to create a creative interface for the AV Panel for Olli2.0 that builds trust in riders.
Olli is the world’s first co-created, self-driving, electric vehicle, designed to reinvent the future of urban mobility and sustainability. Created in 2016 by LM Industries’ subsidiary, Local Motors, Olli is a 3D-printed, self-driving, smart shuttle, built to change the future of mobility.
I came across the Human-AV Connect challenge in launchforth.io and was interested to explore more as UI/UX for autonomous vehicle was very new to me. I started learning about the requirements and deliverables for the challenge and kickstarted my work.
I designed three different user flow for the following :
- Interior Trust Panel — To bring in trust to the riders,
- Interior Route Panel — To share route information
- Exterior Route Panel — To share the route information while passengers get in .
Problem Statement
The challenge was to design a UI/UX mockups to showcase how riders on a university campus would interact with Olli. The tipping point for adoption of Autonomous Vehicles (AVs) is trust and consistently delivering a better experience than conventional transportation options.
Self-driving vehicles are catching up slowly and many companies like Google, Audi, Tesla already have their AVs on-road. Surveys indicate that although people are fascinated by the idea of self-driving cars, most of them are hesitant to relinquish the control of the vehicle.
“ How we can design an interface that increases riders’ trust on Olli2.0 ?”
RESEARCH
To ideate and design the product, I first had to understand the current AV market and players in the market. I also need to understand the type of customers who would use Olli. Since I did not have access to people who have tried AV in India, I took help of Google to understand the user scenario and the pain points better. I read through the research in AV and how AV can play a major role in transportation in the future.
Why AVs are the next big thing?
Apart from advantages like improving travel efficiency, reducing travel insurance cost, reduce commute time, AV’s can reduce the number of road accidents. “Yes I am not kidding, AVs if made a safe vehicle can reduce road accidents!”
A survey by US National Highway Traffic Safety Administration in 2015 shows that
“90% of Road Accidents are due to Human Errors which include intoxication, misjudgement, fatigue, distraction”
All these risks could be reduced by autonomous vehicle as the sensory technology could potentially perceive the environment better than human senses, seeing farther ahead, better in poor visibility, detecting smaller and more subtle obstacles, more reasons for less traffic accidents.
The current research in AV is largely focused on technical aspects and safety but from customer perspective a lot of concerns still exist in terms of trust.
BUILDING TRUST IN AUTOMATION
Research (reference below) suggest that trust in automation develops in four stages
- Information Assimilation
- Trust Evolution
- Intention Formation
- Reliance Action
While interacting with the system, trust is gained by the performance of the system, design feature of the interface and interaction experience.
What are the different ways to build trust?
The research by McKnight and Chervany shows that trust in AVs can be increases by following :
- Possibility to intervene and take manual control
- Anthropomorphism (What!! What does that even mean?)
- System transparency
- Polite Communication with Riders
a) Possibility to Intervene
The AVs interface should give the user a way to take immediate control when the user feels so. Switching between manual control or automation driving mode should be seamless and well communicated via interface. In Audi Self-driving card when user can shift to auto pilot mode, two buttons on the center hub of the steering wheel blinks: Press to engage. Those two buttons were inspired by the famous missile launching systems where two keys had to be turned at the same time, to avoid mistakes.
b)Anthropomorphism
Anthropomorphism is the attribution of human traits, emotions, or intentions to non-human entities. It is considered to be an innate tendency of human psychology.
Research proves that having a human-like character to a non-human entity can build trust among the users.
Where a survey was conducted testing the level of trust among users with Graphical interface and Conversational anthropomorphised interface, the results showed that the people liked and trusted the human-like interface more which had voice, gender , name etc.
Google and Alexa are great examples for this. Even family robots like Zembo, Aido, Pillo all have anthropomorphic interface to interact with the users.
On the other side too much human-likeness is likely to cause a averse effect by creating unrealistic expectation and when system fails to deliver , this could make the users feel the system is flawed.
Apart from having a human-like character , the system interface should also be able to show human-like behaviour in terms of decision making and show proper understanding of the situation and take required action.
c) Transparency
Bringing in transparency in the interface is also a way to gain users trust. Some ways to do this would be to inform the user in advance about the road condition, inform or warn user about any event that is about to take place before it happens. For example, the system can inform if there is speed breaker or a signal in advance . The interface can also show maps or trials taken by the AV to make user aware of what is happening.
d) Pleasant Voice Interface
In order to make the system more believable, the system should have an identity like name, gender , face and most important voice. The genera conversation principle is called the cooperative principle:
“Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged”
IDEATION
In order to design the interface for Olli 2.0, I am considering 3 important features to build trust in the riders and to provide a better , easy to use interface.
- Anthropomorphism
- Transparency
- Voice Interface
Also to design the user interface and user flow, we need to understand the user goals and needs. So my next step was to create user persona and understanding their behaviour.
User Analysis and Persona Mapping
Analysing our target market showed that our users (riders) from the university could be of diverse age from students to staff. Olli 2.0 was setup in university to take people around the campus , so I defined two personas of different age group. I also considered goals, activities, likes and dislikes of the personas.
The persona identification process influenced the following :
1)Based on rider’s interests , like and dislikes Olli can provide extra information while riding like cafe nearby, information regarding any student gathering in the campus for students etc.
2) Based on age and disabilities we need to consider accessibility norms and make sure our UI meets them. for ex, we should also consider color blindness because 8% male and 0.5% female are color blind.
3) Familiarity with mobile phone usage — We need make the UI/UX more simple and straightforward.
4) Ways to integrate the route information in the mobile phone app.
Storyboard
I started the story boarding and defined scenarios and how Olli would react for the same. I also showcased how I use the above mentioned features Anthropomorphism, Transparency and Voice interface to bring in a more trust .
The following screens show how Olli switches the display between human-like face and Route map in the exterios panel. If also display warnings like “Doors Opening” “Doors Closing in the display for riders to be careful.
The below scenes show how the trust panel provides anthropomorphism, transparency in information and a pleasant voice interface to build trust amongst the riders. Olli also engages in conversation with the riders and provides campus related information.
The below screens show how Olli can communicate with other Ollis and optimize route, space and pass information. Incase of break down or any event, other Olli is informed and comes for rescue.
PROTOTYPE
I moved on to creating wireframes and high-fidelity prototypes. I also created Olli, which was fun :)
Wireframes
Olli is here
I also created the anthropomorphic image of Olli, which is displayed in exterior Route Panel and Interior Trust panel from time to time when Olli engages in conversation.
High Fidelity Prototype
I then moved on the creating the high-fidelity prototypes based on the storyboard. I used a simple , feel good design. The panel should have all the necessary information at the same time , the panel should not look over crowded.
Outer route Panel
The outer panel contains
- Route List
- Olli’s Facial Gesture when we try to have a conversation with Olli
- Map that shows the route covered
- Any additional information displayed in the form of cards if asked for.
Inner Trust Panel
The trust panel is placed inside Olli and it displays the following :
- Map that shows the current position of Olli
- Other Vehicles in red and pedestrians in the surroundings in Grey dots.
- Olli also displays the messages on top in occurrence of any event like “Giving way” “Speeding up” “Slowing down” …
Inner Route Panel
The inner route panel is similar to the one that is outside
Apart from displaying the route map, it also displays warnings when the stop has arrive or the door opens/closes.
Reflect and Learn
The whole UI/UX design was a part of a challenge and I had a great learning experience. I pushed myself to complete the whole project in 3 days. I am also looking forward to learn and design interfaces for more such advanced technologies.
What I plan to do next?
The next steps would be to design a mobile app, that allows user to interact with Olli, check Olli timings on mobile and request Olli for pickup. This data will enable Olli to optimise the routes based on requests. Olli App can also be designed to act incase of emergency, where Olli will be sent immediately to the spot for pickup.
If you like my work , feel free to like my work and also ping me if you would like to discuss more.
References : McKnight, D.H.; Chervany, N.L. Trust and distrust definitions: One bite at a time. In Trust in Cyber-Societies; Springer: New York, NY, USA, 2001; pp. 27–54.
Lee, J.D.; See, K.A. Trust in automation: Designing for appropriate reliance. Hum. Factors 2004, 46, 50–80.
