Projects

Ongoing

NRF

World-Model-Based Driving Skill Learning for Humanoid Robots and Real-Vehicle Validation in a V2X Testbed

Personal NRF · Sep 2026 – Aug 2028

A personal NRF-funded project that I lead as principal investigator. Rather than making the vehicle intelligent, it replaces the driver: a humanoid robot operates the steering wheel and pedals of an ordinary car. The driving policy is learned through world-model–based reinforcement learning in a coupled robot–vehicle simulation, then transferred to a real humanoid and validated on an actual vehicle at the KAIST V2X testbed, where the robot itself acts on the infrastructure's cooperative messages.

World Model Humanoid Driving Reinforcement Learning V2X
MOLIT

Development of Virtual Environment and Demonstration Technology for Automated Driving based on Metaverse

Lab MOLIT · 2023 – 2027

A long-running national research program building a metaverse-based virtual environment for validating automated-driving technology. My work in this project focuses on behaviorally realistic traffic agents (pedestrians, cyclists, and other vehicles) that can stress- test autonomous-vehicle decision-making in scenarios that are impractical to reproduce on real roads. This project provides the simulation infrastructure for several of my ongoing studies on reinforcement-learning–based adversarial agents and humanoid-driven vehicle interactions.

Metaverse Automated Driving Virtual Environment Simulation

Past

MOLIT

ASEAN–Korea Transport Cooperation Roadmap (2026–2030)

Lab MOLIT · 2024 – 2025

Working with the ASEAN Secretariat, I led the planning of the ASEAN–Korea transport cooperation roadmap for 2026–2030, identifying five-year priorities across digitalization of public transport, last-mile delivery, and capacity-building among ASEAN member states. The roadmap was presented and discussed at the ASEAN Land Transport Working Group meeting in Cambodia in August 2025.

ASEAN Transport Policy Digitalization International Cooperation
NRF

Developing an Aggressive Pedestrian Model for Autonomous Vehicle Safety Testing

Personal NRF · Jul 2024 – Feb 2025

A personal NRF-funded project to design and validate an "aggressive pedestrian" behavior model for stress-testing autonomous vehicle safety. Most existing pedestrian models in AV simulators are conservative and rule-following; this project instead built a behavioral model that explicitly captures risk-taking and rule-violating crossing patterns, providing a sharper test of an AV's response to vulnerable road users.

Autonomous Vehicles Pedestrian Modeling Safety Testing Adversarial Behavior

* Concluded early upon advancement to the Ph.D. program.