Embedded Files
Smart Autonomous and Infrastructure Lab (SAIL) addresses major challenges in the domains of autonomous driving, Intelligent Transportation Systems (ITS), Urban Air Mobility (UAM), and pedestrian safety, all within the broader landscape of smart mobility and transportation. Our lab actively adopts cutting-edge technologies such as deep learning, generative AI, and multimodal learning to conduct research in fields including perception and control for autonomous vehicles, traffic flow optimization, pedestrian safety improvement, and UAM-based transportation solutions.
In particular, we analyze diverse traffic and urban environment data to tackle real-world problems, such as preventing traffic accidents, predicting vehicle trajectories, and alleviating congestion. We also aim to build efficient and sustainable mobility systems in urban areas through advanced AI and data analytics.
We welcome dedicated and skilled individuals, including undergraduate researchers, Master’s and doctoral students, as well as postdoctoral researchers, to become part of our team. If you are interested in contributing to our research and further developing your expertise, please contact Professor Byeongjoon Noh at powernoh@sch.ac.kr
학부연구생 및 대학원 진학을 희망하는 학생들의 많은 관심 바랍니다. (E-mail: powernoh@sch.ac.kr)
Announcement
(2025) A paper has been accepted for presentation EPIA 2025 (AITS track), Faro, Portugal
Title: "Driving Scene Context-Augmented Trajectory Prediction with Risk-Aware Decision Reasoning using Multimodal LLM"
(2025) A paper has been accepted for presentation at IEEE ITSC 2025, Gold Coast, Australia
Title: "Traffic Context-Augmented Vehicle Trajectory Prediction Framework using Multimodal LLM"
(2025) Our team has achieved state-of-the-art performance in person re-identification problem. [arxiv]
(2025) Our team has achieved state-of-the-art performance in out-of-distribution (OoD) segmentation problem. [arxiv]
(2025) Byeonghun Kim (M.S.) has been appointed at KIST. Congratulations!
(2025) A paper has been accepted for publication in Artificial Intelligence and Law (Top 10%)
Title: "TRACS-LLM: LLM-based Traffic Accident Criminal Sentencing Prediction Focusing on Imprisonment, Probation, and Fines"
(2025) A paper has been accepted for publication in IEEE Access.
Title: "Attention-Driven Lane Change Trajectory Prediction with Traffic Context in Urban Environments"
(2025) Prof. Byeongjoon Noh has been appointed as a member of the Digital Transformation (DX) Policy Advisory Committee of the Korea Transportation Safety Authority (한국교통안전공단). [LINK]
(2025) Prof. Byeongjoon Noh has been appointed as a member of the Advisory Committee of the Daejeon Education Policy Institute (대전교육정책연구소). [LINK]
(2025) A paper has been accepted for publication in Applied Energy (IF=11.0, Top 10%)
Title: "SolarNexus: A deep learning framework for adaptive photovoltaic power generation forecasting and scalable management"
(2025) We have 1 paper accepted in AAAI 2025 Workshop (FLUID), Philadelphia, Pennsylvania, USA. [arXiv]
Title: "FLAMe: Federated Learning with Attention Mechanism using Spatio-Temporal Keypoint Transformers for Pedestrian Fall Detection in Smart Cities"
(2025) We have been selected for the BK21+ Program under the education unit titled "AIoT-에너지 융합기술 지역혁신인재양성 교육단."
(2025) A paper has been accepted for publication in Applied Intelligence.
Title: "Federated learning-based road surveillance system in distributed CCTV environment: Pedestrian fall recognition using spatio-temporal attention networks"
(2025) A paper has been accepted for publication in IEEE Access.
Title: "SaliencyMix+: Noise-Minimized Image Mixing Method with Saliency Map in Data Augmentation"
(2024) We have 1 paper accepted in 104th Transportation Research Board Annual Meeting (TRBAM) 2025, Washington D.C., USA
Title: "DARK-LLM: Description Data Auto-Labeling with Reliable Keyword Using Large Language Model for Analyzing Causes of Autonomous Vehicle Disengagements"
(2024) Jeonghoon Song’s paper was selected as the Best Paper at the 2024 Fall Conference of the Korean Society of Intelligent Transportation Systems.
Title: "지능형 교통 시스템에서 영상 기반 도로 객체 탐지 능력 강화를 위한 Mask2Former와 SAM기반의 Out-of-Distribution 객체 탐지 연구"
(2024) Byeonghun Kim (M.S. course) was selected for the Graduate Student Research Fellowship Program (석사연구지원장려사업) funded by the National Research Foundation of Korea (NRF).
과제명: "차세대 지능형 교통안전시스템을 위한 컴퓨터 비전기반 도로위험행동 예측모델의 개인화 준지도 연합학습 연구"
(2024) We have been launched a new university-industry collaboration project with Nota AI.
과제명: "교차로 내 보행자, 차량 객체 위치 예측 기술 개발"
(2024) A new research project funded by Institute for Information & communication Technology Planning & evaluation (IITP, 정보통신기획평가원) has been awarded.
과제명: "광역권 도시를 위한 차세대 AI융합 모빌리티 시뮬레이션 및 예측/활용 기술 개발"
(2024) A paper has been accepted for publication in Cities (Top 5%)
Title: "Do enhanced school zone policies improve pedestrians' safety? A deep learning-based case study of Osan City, South Korea"
(2024) Sunghun Kim’s paper was selected as the Best Paper at the Annual Conference of KIPS 2024.
Title "미세조정된 VideoLLaMA2 기반의 멀티모달 보행자 횡단 의도 예측"
(2024) A paper has been accepted for publication in Accident Analysis and Prevention (Top 5%)
Title: "Integrating visual and community environments in a motorcycle crash and casualty estimation"
(2024) Papers by Seokjun Hong, Jeonghoon Song, and Hyunsik Min were selected as Best Papers at the 70th Summer Conference of the Korea Computer and Information Science Society.
Title:"DDSP-SVC를 활용한 AI 음성 변환 모델" (Jeonghoon Song)
Title:"차량 내 방송 매체의 은어·속어 감지 및 필터링 시스템" (Seokjun Hong)
Title: "도시 환경 내 교통소음 인식 및 분류를 위한 모델 개발" (Hyunsik Min)
(2024) We are pleased to share that the following undergraduate students from our lab have been selected for the 2024 ETRI Undergraduate Internship Program: Joobin Jin, Jeonghoon Song, Sunghun Kim, Jinseong Kim, Jaegyun Lim, Howoo Lee
(2024) A paper has been accepted for publication in Advances in Engineering Software.
Title: "A novel approach for reliable pedestrian trajectory collection with behavior-based trajectory reconstruction for urban surveillance systems"
(2024) Joobin Jin’s paper was selected as an Best Paper (General Oral Session) at the 2024 Spring Conference of the Korean Society of Intelligent Transport Systems.
Title: "데이터 큐브 모델을 활용한 상업용 차량 위험운전 행동 다차원 분석"
(2024) We have 1 paper accepted in Environmental Design Research Association (EDRA) 55.
Title: "Towards Human-Centered Urban Safety: Designing Urban Environments to Prevent Motorcycle Accidents"
(2024) A paper by Byeonghun Kim has been accepted for publication in the Journal of Advanced Transportation.
Title: "Method on Efficient Operation of Multiple Models for Vision‐Based In‐Flight Risky Behavior Recognition in UAM Safety and Security"
(2024) A paper has been accepted for publication in Applied Geography.
Title: "How do crosswalk delays affect pedestrian access in zoning areas? Walking access reduction by signalized crosswalks in Seoul, South Korea"
(2023) Joobin Jin’s paper was selected as an Best Paper (General Oral Session) at the 2023 Fall Conference of the Korean Society of Intelligent Transport Systems.
Title: "데이터 큐브 모델을 활용한 스쿨존에서의 교통 사고 분석"
(2023) We are pleased to announce the launch of the Smart Autonomous & Infrastructure Lab (SAIL) in the Department of AI and Big Data at Soonchunhyang University.
Page updated
Google Sites
Report abuse