AR Accident Simulation
Accident Simulator is an AR product designing at Phoenix On the Ground(POG) to prevent traffic accidents by simulating frequently happening accidents between cars and pedestrians. The application works based on deep learning data that perceive lanes.
I am working as an AR programmer and product designer to build a safe city together with a deep learning engineer and a project manager. We are continuously developing our content through user testing and researching the effectiveness of the app in decreasing future traffic accidents.
PROJECT Timeline
2019 – Current
Category
Augmented Reality
User Research
Responsibilities
AR Programming
Interface Design
3D Character Design
Research Planning
02. Solution
We designed geo-location-based AR app and 360 videos to educate the drivers and pedestrians about traffic accidents.
The experience brought other’s traffic accidents to an individual level, impacting users to prevent the accident by reminding the implanted memories.
AR contentsv(10 CASES)
Developed AR app simulating
accidents between pedestrian and driver
Integrated deep learning to
detect the lanes
360 videov(30 CASES)
Developed 360 video version to supplementation issues in AR
Created separate content with ‘driver point of view’ and ‘pedestrian point of view’ to effectively communicate how accidents occur
03. Result
The project successfully got funded by multiple public and private corporations and we are aiming to launch the app to the public.
Fundings
Korea Expressway Corporation
Korea Institute of Civil Engineering and Building Technology (KICT)
Yonsei University Techstars
Goyang K-startup
QR CODE device
QR code device installed on few locations for further user testing
04. Process
For this project, I joined the team in the design and prototype phase. Business concepts and plans were built by the team leader.
The process I have not participated in is marked in grey.
05. Research
Traffic accidents are increasing year by year, showing a decrease in kill figures and an increase in injured figures. It is important to figure out the focal reason for accidents and identify the different patterns. We found drivers’ and pedestrians’ failure to keep their eyes on the road is the main issue, which is less recognized.
The user-based insurance (UBI) market is increasing. Most of the companies reflect driver’s behavior on the bill, such as vehicle miles and acceleration/ decelerationHowever, those UBI strategies only focus on the cars, not focusing on how different actions drivers take.
06. Opportunity
“Planting memory of near-miss traffic accident”
We found developing the ability to prepare for similar accidents by instilling memories of accidents is the most effective solution to the main cause of traffic accidents.
07. Design & develop
character DESIGN
Design a 3D model for simulation without traumatic influence
Develop, phase1
Create AR app and integrate deep learning for precise adjustment
DEVELOP, PHASE2
Create 360 video and design an app to publicize the service
USER SCENARIO DESIGN
The traffic accident simulation happens at the site where accidents are prone to happen. The user uses one’s phone and goes through vicarious experience.
Considering the context, we aimed to design a safe service, reminding users of the surrounding through visual and audio cues.
A. The user finds the POG marker on the street and scans the QR code to start the AR experience that fits the corresponding location. The marker alerts the user of the surroundings and notifies them once the traffic signal changes.
B. The user follows the instruction on the AR app. The guideline allows the user to stand at the secure position and point the camera at the proper angle.
C. The user learns about the accident frequently happening on the location through AR simulation.
D. The user checks out the information on the injury and a written description of the cause of the accident.
3D character designed in Fusion 360
Nodding and looking around motion applied in Unity
develop phase 1
AR DEVELOPMENT
With the designed character and user scenario, we developed an AR app, coming up with ten different cases demonstrating frequently happening accidents.
AR simulation on the accident between car and pedestrian
DEVELOP PHASE 1
DEEP LEARNING INTEGRATION
The service aims to provide a seamless user experience where the simulation contents are augmented at where it should be, especially vehicles.
I worked with a deep learning engineer who trained a deep learning model to differentiate between lanes and not-lanes and converted the data into coordination usable at Unity. I was responsible for using those data and augment the simulation at the right location on the screen.
Scene to train
Trained result showing lane in white and non-lane in black
08. Evaluate and result
Selected for the Scale-Up project from the Ministry of Land, Infrastructure and Transport, we got funded to conduct user testing to find out the effectiveness of the AR accident simulation.
Brainwave & eye tracking
Tested response, trauma stability, perception of reaching risk area
A participant going through user testing
TEST procedure
USER TEST RESULT
“AR traffic accident simulation system is a psychologically stable training method and shows a significant impact on preventing incidents”
(Eye tracking)
Field distribution adequacy
The ability to respond to accidents
increased by 69.7 percent
(Brainwave)
Trauma stability
AR simulation is safer than educating with
real accident video by 14.8 percent
(Brainwave)
Perception of reaching risk area
On the second week,
increased by 43.3 percent
09. Further development
After we proved the effectiveness of augmented traffic accidents education, we expanded the scope of traffic accident simulation into three different types of perspectives.
