Hyundai AUI
Acoustic UI Concept Development
Introduction
As more technology is put into cars, drivers are enriched with more and more functions, and so have increased distractions. The distraction arises because manufacturers simply focused on providing more functions yet using the traditional buttons and touch screens. Hyundai researchers have teamed up with KAIST (Korea Advanced Institute of Science and Technology) researchers and came up with a sensor that can be used as an input device for recognition of touching and scrubbing, and they have requested us to create an acoustic based user interface concepts that utilizes this input method.
Technology
The technology utilizes rather simple sensors and patterned plates. When the plate is touched or rubbed, its patterned surface would create different sounds depending on the action, which are then picked up by the sensor. The algorithm interprets these sounds and translates if it’s a tap or slide, a thump, or even the directions of the rubbing. A controller using this technology can be applied on any surface of a vehicle interior, preferably on a steering wheel or center fascia
(via patent KR101558810B1).
Research
We began breaking down the AUI technology and the vehicle functions to understand the lowest level of elements and to see how we could remap these as a working user interface.
Ideation
We processed the raw data that was gathered from the research to create comprehensive UI sets. Then we improved them through multiple idea sessions until the two final sets have evolved.
AUI Set A: Two Hand Model
The two hand UI model utilizes two hands of a driver and a structured menu. The advantage of this UI set is that it can be readily applied to the current infotainment structure. The use of left and right hands are directional, and it’s easily mapped to the user’s mental model of a controller.
Requirements
AUI Input Definition
Flow Chart
AUI Set B: One Hand Model
The one hand UI model uses combination of inputs. In this way, the user can execute functions without scrolling through a menu. The problem with this model is that the user has to remember the commands, and initially it takes more time to learn the commands because it’s a completely new mental model. However, once you know the commands, it’s faster and less distracting to use while driving.
Requirements
