Research

I'm in the first year of my M.S. and working toward research on automated vehicles and mobility services in mixed traffic. Right now I'm reading the literature, learning traffic simulators such as SUMO and VISSIM, and contributing to the lab's national R&D projects below.

Interests

Connected & automated vehicles in mixed traffic

Vehicle-level prediction and planning when automated vehicles share the road with human drivers, infrastructure-aware and V2X-enabled planning, cooperative driving, and multi-agent interaction.
- AI: learned behavior prediction, reinforcement learning and world models for planning.
Shared autonomous mobility & robotaxi operations

Matching and dispatching, routing, fleet rebalancing, ride-pooling, and service planning for robotaxi and shared mobility fleets.
- AI: reinforcement learning and learning-based optimization for fleet operations, demand forecasting.
Travel behavior & system-level evaluation

How people choose robotaxis and automated vehicles (mode choice, adoption), value of time, willingness to pay, and elasticities, and what new mobility does to congestion, VKT, and transit.
- AI: LLM agents as synthetic travelers, and checking whether they respond to changing conditions the way people do, not only whether they predict choices.

National R&D projects

Physical AI and Infrastructure-Centered AV Control Framework for Heterogeneous Mixed Autonomous Traffic

Aug 2026–present

National Research Foundation of Korea (NRF) · Participating researcher

The project studies how automated vehicles can predict surrounding agents and decide their own actions at intersections where they mix with human-driven vehicles.
My role: literature review.

World Foundation Model-Based E2E Autonomous Driving and Real-Vehicle Global Demonstration

Jul 2026–present

Korea Technology and Information Promotion Agency for SMEs (TIPA), Global TIPS · Participating researcher

My role: related research and paper writing.

Metaverse-Based Virtual Testing Environment and Demonstration Technology for Autonomous Driving

May 2026–present

Korea Agency for Infrastructure Technology Advancement (KAIA) · Participating researcher

Joined in the final phase of the project; I manage living-lab demonstration data.

Past projects

AI-Based Demand Modeling for Autonomous Ride-Hailing

Sep 2025–Jul 2026

Yonsei capstone, continued at KAIST

Proposed an AI-based framework (persona simulation) for forecasting demand for autonomous ride-hailing and built it as a multi-agent system.
Finished the research at KAIST with an evaluation framework, as joint Yonsei–KAIST work; submitted to Transportation Research Part C (under review).
Grand Prize, ST–Yonsei Capstone Design; Korean Society of Transportation President's Award for top capstone.

Marketing Solution Agent for Small Merchants13th BigContest

Sep–Dec 2025

AI Data Utilization Track · Best Award, KAIT President's Award (2nd of 500+ teams)

Built a multi-agent LLM system that diagnoses each merchant's situation from card-sales, foot-traffic, and street-view data and recommends tailored marketing strategies.

RAG Agent for Counseling-Record AnalysisUpstage

May–Jun 2025

Industry–academia project with Upstage

Parsed and anonymized scanned counseling-record PDFs, retrieved similar cases, and classified risk level and abuse type with an LLM agent (Upstage Document Parse, Solar LLM, FAISS).
Added automatic alerts for high-risk cases and a Streamlit monitoring dashboard.

Namdaemun Market Night-Time Revitalization12th BigContest

Sep–Dec 2024

Data Analysis Track · Best Award, KAIT President's Award (2nd of 200+ teams)

Used SKT mobility (OD), commercial, and parking data to find the problem (visitors leaving in the evening), trace its causes (early closing, parking shortage), and propose solutions (a night-market district and new parking sites).