ADII (Autonomous Driving Intervening Interface) is a physical interface that allows to intervene level 5 autonomous driving.

Automation of cars will bring vast changes to many aspects of cars. Autonomous cars are not likely to be owned by individuals and interiors of autonomous cars will be shaped differently from the cars of the past 130 years. Many constraints of current car interface design will not matter in autonomous cars. For example, interfaces in cars don't need to be around the center console since there may not be a designated driver's seat. The primary activity of passengers in the autonomous cars is going to be non-driving related tasks and drivers will engage in driving only when they want to. These changes will require designers to design physical interfaces according to the different user experience. Different forms of physical interfaces for driving in autonomous cars can be explored for the different driving experience in autonomous cars.

ADII (Autonomous Driving Intervening Interface) is designed to answer following questions.

How should physical interfaces designed according to the different user experience proposed by autonomous cars?

How could physical interface design strengthen the autonomous car experience?

RESEARCH

ADII's design process involved secondary researches about advantages of physical interfaces and changes proposed by autonomous cars.

- Advantages of physical interfaces

Physical interfaces can provide tactile information which can amplify the humans' ability of sensing. Due to our ability of sensing, we can subconsciously use physical interfaces effortlessly. When we are using physical interfaces we can subconsciously collect numerous pieces of information. For example, while we are using a volume control knob adjustment of volume through haptic feedback. Also, we can recognize the spectrum of volume when we can't turn the knob further.

- Changes proposed by autonomous cars

In theory level 5 autonomous cars won't need driver's intervention since it provides fully safe self-driving under all circumstances The primary activity of passengers in the autonomous cars is going to be non-driving related tasks. Nonetheless, passengers still may want to engage in driving occasionally. Many constraints of current car interface and interior designs will not matter in autonomous cars. For example, seats can be reconfigured to face different directions. Maneuvers for providing Input for intervening autonomous driving won't be fine-grained control level. This maneuvers will only be done from time to time and drivers will not be trained as extensive as they currently are.

Semi-structured interviews with a car interior designer and a HMI (Human Machine Interface) designer in the automotive industry were conducted in order to learn about the physical interface design opportunities in autonomous cars. These are the key insights I learned from the interviews:

The main challenge of interface design in the future cars would be the communication between a driver and a car, for example how to tell the user to take back the control at the key moment, how to express visual control metaphors for autonomous features.

The context of using the interface needs to be considered, whether the interface is used in an emergency or situation that is unrelated to driving.

When there is a newly designed interface that changes cognitive behavior, many things need to be considered. It is important not to violate the user's expectations.

PROCESS

Based on the insights I gained from the secondary research and the interviews with experts, I came up with various physical interface design concepts for autonomous cars. I interviewed 6 interviewees to validate and improve the concepts and choose the final design. I briefed them about the key insights I learned from the researches and explained how the concepts work. For each concept, I asked interviewees what are the things they think that are working well for the design and their concerns and suggestions about the concept. Final design was chosen based on the feedbacks from the interviewees. Autonomous car shaped remote with 3D interactive navigation map concept was considered as the strongest and most interesting concept among the interviewees, they liked the idea of moving something physically. For the final design, I created physical prototypes to study scale and physicality of the concept.

CONCEPT

ADII (Autonomous Driving Intervening Interface) is composed of two main components: the autonomous car shaped remote with physical buttons (ADII remote) and the 3D interactive navigation map on the Interactive screen table inside an autonomous vehicle interior. Interactive screen table displays realistic vivid 3D navigation map that passenger can interact with the ADII remote (ADII remote attaches on the interactive screen by magnetic force).

ADII remote is shaped as an autonomous car in order to convey visual cues and leverage familiar experiences. Interacting with the physical car shaped remote with buttons on the vivid 3D map provides intuitive and direct way of controlling as if you are using a computer mouse or playing with Hot Wheels.

Having the Interactive screen table that displays informative 3D navigation map inside an autonomous car between seats provides a prompt way of intervening level 5 autonomous driving for the passengers who were engaging in non-driving related tasks. Passengers can direct the autonomous car to the exact route that they would like to take to the destination.