Users Report / Vol.153
The first and only government–academia research institute in Korea, jointly established by Gyeonggi-do Province and Seoul National University, aims to strengthen national competitiveness by enhancing integrated science and technology innovation and regional scientific capabilities.
Location: Iui-dong, Suwon-si, Gyeonggi-do Province, Korea
URL: https://aict.snu.ac.kr/kor.do
Director of Intelligent Transport Systems (ITS) Laboratory / Team Leader of Gyeonggi-do Province Autonomous Driving Center
Ph.D Hyungjoo Kim
Leading Korea’s Next-Generation Mobility Research Through Both Scientific Study and Practical Implementation
Developing an Experimental Platform for Real-Time Visualization of Autonomous Driving Risk Prediction Using UC-win/Road
The Advanced Institute of Convergence Technology, jointly established in 2005 by Gyeonggi-do Province and Seoul National University, is Korea’s first government–academia research institute. It promotes convergence research across information technology, biotechnology, nanotechnology, and related fields. It comprises several specialized laboratories dedicated to developing next‑generation industries, among which the Intelligent Transport Systems (ITS) Laboratory is responsible for the autonomous driving field. Ph.D Hyungjoo, the director of the laboratory, leads demonstration research on autonomous vehicles and infrastructure technologies and also operates a large-scale test bed.
In addition to performing demonstration researches of autonomous vehicle technology development and provides actual autonomous driving services, the Advanced Institute of Convergence Technology has operated the integrated management center for seven years. It collects data on autonomous vehicles and infrastructure generated from the test bed, analyzes and processes it, and provides it to external parties.
A test environment using real roads has been established, serving as a base that links research with implementation. Its center hub is Pangyo Techno Valley in Seongnam, Gyeonggi-do Province. Located about 20 km south of central Seoul, the area is an IT and R&D hub developed to become Korea’s Silicon Valley. It has now developed into a leading hub where many autonomous‑driving companies and startups gather.
In the second district of Pangyo Techno Valley, an open‑platform autonomous driving demonstration complex was developed between 2016 and 2021. With the completion of the autonomous driving infrastructure, Gyeonggi Autonomous Driving Center (GADC) was established as a dedicated organization for its operation and management. The center serves as the managing organization for Pangyo Zero City, which has been designated as an autonomous driving demonstration area, and Prof. Kim also works there as the leader of the autonomous driving R&D team.
GADC supports demonstration tests and collects and manages driving big data through the operation of its integrated management center and data center, as well as the autonomous‑driving infrastructure—such as IoT facilities and V2X systems—installed in the first and second Pangyo Techno Valley districts. It also develops test environments for autonomous vehicles and sensors on public roads, providing a framework that enables demonstrations in everyday settings.
As part of its support for autonomous‑driving startups, it promotes an industrial ecosystem by providing research spaces, running incubation programs, and supporting vehicle development and commercialization, and it also operates Korea’s first public autonomous vehicle, the Zero Shuttle.
The Zero Shuttle operates as a demonstration vehicle connecting the first and second districts of Pangyo Techno Valley, serving as a platform for technology testing and big‑data collection. In 2022, the demonstration area was expanded from the first to the second district of Pangyo Techno Valley, creating an environment where users can request autonomous‑driving services from anywhere in the area. This has laid the foundation for gradually transitioning from technology demonstrations to real‑world services.
Working across both organizations, he drives technological development in Korea’s next‑generation mobility policy from both the research and implementation sides.
Road environment monitoring service
It is installed on the bridge connecting Zone 1 and Zone 2 of the second Pangyo Techno Valley and collects and provides road‑surface data based on weather conditions.
Signal display service
It is installed at signalized sections in Pangyo Zero City and provides signal‑phase information along vehicle driving routes.
R&D on autonomous‑driving technologies is carried out with government support. Over the past two years, he has been engaged in a joint project with the police to build a platform capable of predicting accidents in real time. The project "SILS for Real-time Risk Prediction Model Verification based on UC-win/Road" that received the Excellent Award of the 24th 3DVR Simulation Contest on Cloud in 2025 is the result of collaborative research with the police.
Through UC-win/Road’s real-time interface, data such as position, speed, acceleration, and brake signals are streamed to external prediction models, allowing real-time visualization of risk forecasts. According to Prof. Kim, this platform can assess the risk level of driving behaviors and individual vehicles in real time and serves as a highly effective tool for preventing hazards during vehicle interactions.
Prof. Kim first encountered UC-win/Road while studying traffic engineering as a university student.
UC-win/Road is widely recognized in the Korean traffic industry, and he learned about it through a professor specializing in traffic safety. When he tried it, he found that it allows for various simulations needed to address traffic‑safety challenges. He is pleased to have found software with such powerful simulation capabilities.
Robust simulations are essential for building platforms that evaluate real‑time risk‑prediction models. He noted that UC-win/Road's highly flexible scenario function enables the reproduction of a wide range of complex situations.
He developed a plugin using the SDK to automatically generate 200 scenario patterns, which greatly improved the efficiency of digital‑twin creation and contributed significantly to his experiments. Given UC-win/Road's expandability, he plans to further advance the current platform and enhance the sophistication of its simulations.
The 24th 3DVR Simulation Contest on Cloud 2nd Prize (Excellent Award)
SILS for Real-time Risk Prediction Model Verification based on UC-win/Road
A Software-in-the-Loop Simulation (SILS) was developed to enable safe and efficient validation of autonomous driving algorithms and real-time risk prediction models.
Through UC-win/Road’s real-time interface, data such as position, speed, acceleration, and brake signals are streamed to external prediction models, allowing real-time visualization of risk forecasts. By automating batch simulations and data collection, the system supports reproducible experiments and quantitative evaluations (e.g., scenario coverage and risk scoring accuracy). The platform shifts initial validation to a virtual environment, reducing both cost and safety risks. It accelerates algorithm iterations while providing a reproducible workflow for testing autonomous driving and driver assistance functions within UC-win/Road.
