PNU / Electronic Engineering / Est. 2026

Beyond the
five senses.
Extended physical
intelligence.

We study Physical Intelligence — perception, decision-making, and action grounded in the laws of physics, not data alone. Our systems are built to perceive, reason, and act reliably in complex real-world settings, across Autonomous Vehicles, Robotics, and Industrial AI.

News

SEP 2026 Conf

A paper accepted to ECCV 2026.

Details →

OCT 2025 Conf

Two papers accepted to IROS 2025.

Details →

JUN 2025 Conf

A paper accepted to IEEE IV 2025.

Details →

FEB 2025 Journal

A paper accepted to IEEE T-ITS.

Details →

OCT 2024 Journal

A paper accepted to IEEE T-CPMT.

Details →

JUL 2023 Journal

A paper accepted to IEEE IoT-J.

Details →

Topics

→ THREAD 01

Autonomous Vehicle & Robotics

Building autonomous systems that perceive and navigate the toughest real-world environments, including what hides behind walls and around corners. We read radar, acoustic, and multi-modal reflections to localize what direct sensors cannot see, and act on it safely.

NLoS Perception Multi-modal Sensing Extended Perception Decision Making

1 journal · 8 conferences · 2 patents → view publications

→ THREAD 02

Smart Manufacturing & AI Systems

Closing the loop between physical processes and AI on the factory floor. We detect rare defects that traditional quality control misses, optimize complex production trade-offs, and pair digital twins with explainable models so operators can trust and act on every result.

Anomaly Detection Digital Twin Explainable AI Pareto Optimal

3 journals → view publications

→ THREAD 03

Real-to-Sim & Data Synthesis

How do we synthesize reality faithfully enough that what trains in simulation survives in the real world? We study sensor and data synthesis, domain randomization, and real-to-sim-to-real transfer. We build physical simulators where physics holds and learned models cross the sim-to-real gap.

Data Synthesis Domain Randomization Physics-based Modeling Sim-to-Real

1 journal → view publications

Publications

* equal contribution · † corresponding author

'26

— Blind review —

Jeonghwan Choo, Mingu Jeon†, Seong-Woo Kim†

Submitted to IEEE Transactions on Industrial Informatics (T-II) · UNDER REVIEW

'26

— Blind review —

Minuk Lee, Mingu Jeon†, Seong-Woo Kim†

Submitted to IEEE Transactions on Industrial Informatics (T-II) · UNDER REVIEW

'26

[P2]

Reflection-Aware Reasoning for Non-Line-of-Sight Pedestrian Localization

Seong-Woo Kim, Byeonggyu Park, Hee-Yeun Kim, Mingu Jeon

Korean Patent Application · 10-2026-0166595 · Filed Sept. 2, 2026

'26

[P1]

Method and Apparatus for Estimating a Position of an Object in a Non-Line-of-Sight Region

Byong Hyok Choi, Chul Woo Chun, Soo Mok Lee, Sung Woo Kim, Mingu Jeon, Hee-Yeun Kim, Byeonggyu Park

Korean Patent Application · 10-2026-0148994 · Filed Aug. 10, 2026

'26

[C8]

Reflection-Aware Reasoning for Non-Line-of-Sight Pedestrian Localization

Byeonggyu Park, Mingu Jeon†, Seong-Woo Kim†

European Conference on Computer Vision (ECCV) · Malmö · 2026

BK21

LINK

'25

[C7]

mmWave Radar-Based Non-Line-of-Sight Pedestrian Localization at T-Junctions Utilizing Road Layout Extraction via Camera

Byeonggyu Park, Hee-Yeun Kim, Byonghyok Choi, Hansang Cho, Byungkwan Kim, Soomok Lee, Mingu Jeon†, Seong-Woo Kim†

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · HANGZHOU · 2025

KIISE

LINK

'25

[C6]

Radar-Based Darting Out NLoS Pedestrian Localization Near Parked Vehicles Using Camera-Assisted Point Cloud Interpretation

Hee-Yeun Kim, Byeonggyu Park, Byonghyok Choi, Hansang Cho, Byungkwan Kim, Soomok Lee, Mingu Jeon†, Seung-Woo Seo†, Seong-Woo Kim†

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · HANGZHOU · 2025

KIISE

LINK

'25

[C5]

Quantifying Potential Risk through Passively Sensed Invisible Modalities

Mingu Jeon, Azimov Dilmurod Turdiboevich, Seung-Woo Seo, Seong-Woo Kim

International Technical Conference on Circuits/Systems, Computers, and Communications (ITC-CSCC) · SEOUL · 2025

LINK

'25

[C4]

Non-Line-of-Sight Multi-Target Localization in T-Junctions Using Ray Tracing of mmWave Radar

Mingu Jeon, Byeonggyu Park, Hee-Yeun Kim, Yujeong Kang, Byonghyok Choi, Hansang Cho, Byungkwan Kim, Soomok Lee, Seung-Woo Seo, Seong-Woo Kim

IEEE Intelligent Vehicles Symposium (IV) · CLUJ-NAPOCA · 2025

LINK

'25

[J4]

Non-Line-of-Sight Vehicle Localization based on Sound

Mingu Jeon, Jae-Kyung Cho, Hee-Yeun Kim, Byeonggyu Park, Seung-Woo Seo, Seong-Woo Kim

IEEE Transactions on Intelligent Transportation Systems (T-ITS) · VOL. 26, NO. 2

JCR Top 2.4%

LINK

'25

[D4]

거대언어모델을 활용한 고객 맞춤형 이미지 분류 시스템 개발 연구

추정환, 전민구, 김성우

대한전자공학회 추계종합학술대회 (IEIE Fall Conference) · Nov 2025

'25

[D5]

자동차 전장 인쇄회로기판 실장품의 전원 분기 구조 자동 설계를 위한 CP-SAT 기반 근사해 탐색기

이민욱, 전민구, 김성우

대한전자공학회 추계종합학술대회 (IEIE Fall Conference) · Nov 2025

'24

[J3]

Extremely Rare Anomaly Detection Pipeline in Semiconductor Bonding Process with Digital Twin-driven Data Augmentation Method

Mingu Jeon*, In-Ho Choi*, Seung-Woo Seo, Seong-Woo Kim

IEEE Transactions on Components, Packaging and Manufacturing Technology (T-CPMT) · VOL. 14, NO. 10

LINK

'23

[J2]

Machinery Value Estimation Method based on IIoT System Utilizing 1D-CNN Model for Low Sampling Rate Vibration Signals from MEMS

Gyuho Lee, Seong-Woo Kim, Mingu Jeon†

IEEE Internet of Things Journal (IoT-J) · VOL. 10, NO. 14

JCR Top 4.1%

LINK

'22

[J1]

A Contactless PCBA Defect Detection Method: Convolutional Neural Networks with Thermographic Images

Mingu Jeon*, Siyun Yoo*, Seong-Woo Kim

IEEE Transactions on Components, Packaging and Manufacturing Technology (T-CPMT) · VOL. 12, NO. 3

LINK

'22

[C3]

Finding Second Hierarchy Point for Cooperative Multi-Vehicles using P-median and Voronoi Diagram Vertices

Kyunghyun Kim, Mingu Jeon, Soohyun Shin, Seong-Woo Kim

International Conference on Electronics, Information, and Communication (ICEIC) · JEJU · 2022

LINK

'22

[D3]

깊이 정보를 활용한 자율 주행에서의 객체 추적 기술

전민구, 김성우

대한전자공학회 하계종합학술대회 (IEIE Summer Conference) · Jun 2022

'21

[C2]

Uncertainty-Aware Fast Curb Detection Using Convolutional Networks in Point Clouds

Younghwa Jung, Mingu Jeon, Chan Kim, Seung-Woo Seo, Seong-Woo Kim

IEEE International Conference on Robotics and Automation (ICRA) · XI'AN · 2021

BK21

LINK

'19

[D2]

HTP 테스트를 통한 아동 그림 감성 분석 자동화 사전 연구

박정현, 신승호, 김재엽, 박기현, 이승준, 전민구, 김성우

한국HCI학회 학술대회 (Korea HCI Conference) · Feb 2019

'18

[C1]

Real-Time Multi Object Tracking with Target Class for Autonomous Vehicles

Mingu Jeon, Seong-Woo Kim

International Conference on Energy and Sustainability (ICES) · SEOUL · 2018

'17

[D1]

Deep Neural Network를 이용한 자율 주행 자동차의 Traffic Light Detection

전민구, 서승우

대한전자공학회 하계종합학술대회 (IEIE Summer Conference) · Jun 2017

People

Mingu Jeon, Ph.D.

Principal Investigator · Assistant Professor

Mingu Jeon is an Assistant Professor in the Department of Electronic Engineering at Pusan National University. His research pursues Extended Physical Intelligence, building perception systems that reach beyond the five senses through radar, sound, and physics-grounded AI.

He received his Ph.D. in Electrical and Computer Engineering from Seoul National University in 2025 under the joint supervision of Prof. Seung-Woo Seo and Prof. Seong-Woo Kim, and his B.S. in Electrical Engineering from POSTECH in 2017.

MAILmingujeon@pusan.ac.kr

OFFCCore Research Facilities, 609

WEBGoogle Scholar · LinkedIn

Members

Jiseok Yang

Integrated B.S.–Ph.D.

tasan@pusan.ac.kr

Woo Seok Seo

Integrated B.S.–Ph.D.

oneplus02@pusan.ac.kr

Dongjae Cho

Undergraduate

whehdwo3636@pusan.ac.kr

Jaemin Park

Undergraduate

awoals0325@pusan.ac.kr

Dongjin Lee

Undergraduate

djlee2623@pusan.ac.kr

Collaborative Researchers

Seoul NationalUniversityKR

Institution

Seoul National University

Seong-Woo KimAssociate Professor
Hee-Yeun KimPh.D. Candidate
Byeonggyu ParkPh.D. Candidate
Hanbi BaekPh.D. Candidate
Jeonghoo SeongM.S. Candidate

Kennesaw StateUniversityUS

Institution

Kennesaw State University

Soomok LeeAssistant Professor

Chungnam NationalUniversityKR

Institution

Chungnam National University

Jun KimAssistant Professor

Gachon UniversityKR

Institution

Gachon University

Ilchae JungAssistant Professor

Tap an institution to see its researchers

Projects

Thread 01Autonomous Vehicle & Robotics

'26
—

AI-based Autonomous Terrain Reconnaissance for CBRN Contamination Mapping

with Hanwha Aerospace

UGV/UAV PLATFORM · PATH PLANNING · CHEMICAL CONTAMINANT MAPPING

'23
—'25

Multimodal Non-Line-of-Sight Object Localization

with National Research Foundation of Korea & Samsung Electro-Mechanics

MULTIMODAL FUSION · NLoS LOCALIZATION · AUTONOMOUS DRIVING

PAPER

'23
—'25

mmWave-based Non-Line-of-Sight Object Localization

with Samsung Electro-Mechanics

mmWave RADAR · NLoS LOCALIZATION · AUTONOMOUS DRIVING

PAPER

'22
—'25

Sound-based Non-Line-of-Sight Vehicle Localization

with National Research Foundation of Korea

ACOUSTIC SENSING · NLoS LOCALIZATION · AUTONOMOUS DRIVING

PAPER

'22
—'24

ARIL Dataset Construction — Acoustic Recognition based Invisible-target Localization

with National Research Foundation of Korea & Samsung Electro-Mechanics

ACOUSTIC SENSING · BEV FISHEYE · NLoS DATASET

DATA

Thread 02Smart Manufacturing & AI Systems

'26
—

Securing Bio-Data Governance for National Plant Genetic Resources — K-BDS Enhancement

with National Research Foundation of Korea

ENVELOPE ENCRYPTION · CONSENT-BASED ACL · BIO-DATA GOVERNANCE

'25
—'26

Explainable AI for Quality Control of FPCB

with LG Innotek

THERMOGRAPHIC IMAGING · EXPLAINABLE AI · QUALITY CONTROL

'25
—'26

Transformer-based Optimization Framework in Electronic Circuit Design

with LG Electronics

LEARNING-BASED · PARETO OPTIMIZATION · CIRCUIT DESIGN

'22
—'23

Anomaly Detection of Bonding Robots in Semiconductor Manufacturing

with Samsung Electronics

DIGITAL TWIN · ANOMALY DETECTION · SEMICONDUCTOR MANUFACTURING

PAPER

'22
—'23

Mortgage Valuation through Status Detection of Movable Assets

with KT

MEMS SENSOR · 1D-CNN · ASSET VALUATION

PAPER

'21
—'22

Deep Learning-based Contactless PCBA Defect Detection

with LG Electronics

THERMOGRAPHIC IMAGING · CNN ANOMALY DETECTION · PCBA MANUFACTURING

PAPER

Dataset

Release · NLoS Localization

ARIL Dataset

Acoustic Recognition based Invisible-target Localization

The ARIL dataset supports research on detecting and localizing Non-Line-of-Sight (NLoS) vehicles. It spans variations in velocity (5, 10, 15, 20 km/h), direction of travel (left, right), and spatial layout (T-Junction configurations, with and without an opposite-side wall).

ARIL was collected in a controlled test bed that reproduces real road conditions. To capture the full scene and ground truth in a single view, a 12-megapixel fisheye camera (185° field of view) was mounted 7 m above the intersection center as the Bird's-Eye-View (BEV) camera. An SUV equipped with a microphone array and camera served as the data-collection vehicle, and the NLoS vehicle's location was obtained from the BEV camera. The total dataset size is 11.7 TB.

Each scenario provides one sound sequence, one BEV image sequence, and one .xlsx file with the NLoS vehicle's location. Audio was recorded at 48,000 Hz and BEV images at 10 Hz.

Access Dataset

Teaching

Lecturer @ PNU

Engineering Mathematics II '26 Fall · Dept. of Electronic Engineering
Circuit Theory I '26 Fall · Dept. of Electronic Engineering
Logic Circuits '26 Spring · Dept. of Electronic Engineering

Co-Lecturer @ SNU

Autonomous Robot Intelligence Graduate seminar
Artificial Intelligence for Manufacturing Graduate seminar
Sensors and Actuators Arusha Technical College · Tanzania

Teaching Assistant @ SNU

Semiconductor Manufacturing Data Analysis Advisory SK Hynix
Advanced Digital Fabrication and Practice Graduate course

Invited Talks

Conferences & Symposia

Summer Workshop on Computer Communications (SWCC) 2026 Aug. 24, 2026 · Seoul
The 21st IEMEK Symposium on Embedded Technology (ISET) 2026 May 15, 2026 · Jeju

Institutional Seminars

POSTECH Sept. 11, 2026 · Dept. of Industrial and Management Engineering
LG Electronics Sept. 9, 2026 · Corporate seminar
Korea Institute of Industrial Technology (KITECH) Aug. 25, 2026

Beyond thefive senses.Extended physicalintelligence.

News

Topics

Autonomous Vehicle & Robotics

Smart Manufacturing & AI Systems

Real-to-Sim & Data Synthesis

Publications

— Blind review —

— Blind review —

Reflection-Aware Reasoning for Non-Line-of-Sight Pedestrian Localization

Method and Apparatus for Estimating a Position of an Object in a Non-Line-of-Sight Region

Reflection-Aware Reasoning for Non-Line-of-Sight Pedestrian Localization

mmWave Radar-Based Non-Line-of-Sight Pedestrian Localization at T-Junctions Utilizing Road Layout Extraction via Camera

Radar-Based Darting Out NLoS Pedestrian Localization Near Parked Vehicles Using Camera-Assisted Point Cloud Interpretation

Quantifying Potential Risk through Passively Sensed Invisible Modalities

Non-Line-of-Sight Multi-Target Localization in T-Junctions Using Ray Tracing of mmWave Radar

Non-Line-of-Sight Vehicle Localization based on Sound

거대언어모델을 활용한 고객 맞춤형 이미지 분류 시스템 개발 연구

자동차 전장 인쇄회로기판 실장품의 전원 분기 구조 자동 설계를 위한 CP-SAT 기반 근사해 탐색기

Extremely Rare Anomaly Detection Pipeline in Semiconductor Bonding Process with Digital Twin-driven Data Augmentation Method

Machinery Value Estimation Method based on IIoT System Utilizing 1D-CNN Model for Low Sampling Rate Vibration Signals from MEMS

A Contactless PCBA Defect Detection Method: Convolutional Neural Networks with Thermographic Images

Finding Second Hierarchy Point for Cooperative Multi-Vehicles using P-median and Voronoi Diagram Vertices

깊이 정보를 활용한 자율 주행에서의 객체 추적 기술

Uncertainty-Aware Fast Curb Detection Using Convolutional Networks in Point Clouds

HTP 테스트를 통한 아동 그림 감성 분석 자동화 사전 연구

Real-Time Multi Object Tracking with Target Class for Autonomous Vehicles

Deep Neural Network를 이용한 자율 주행 자동차의 Traffic Light Detection

People

Mingu Jeon, Ph.D.

Jiseok Yang

Woo Seok Seo

Dongjae Cho

Jaemin Park

Dongjin Lee

Projects

Thread 01Autonomous Vehicle & Robotics

AI-based Autonomous Terrain Reconnaissance for CBRN Contamination Mapping

Multimodal Non-Line-of-Sight Object Localization

mmWave-based Non-Line-of-Sight Object Localization

Sound-based Non-Line-of-Sight Vehicle Localization

ARIL Dataset Construction — Acoustic Recognition based Invisible-target Localization

Thread 02Smart Manufacturing & AI Systems

Securing Bio-Data Governance for National Plant Genetic Resources — K-BDS Enhancement

Explainable AI for Quality Control of FPCB

Transformer-based Optimization Framework in Electronic Circuit Design

Anomaly Detection of Bonding Robots in Semiconductor Manufacturing

Mortgage Valuation through Status Detection of Movable Assets

Deep Learning-based Contactless PCBA Defect Detection

Dataset

ARIL Dataset

Teaching

Lecturer @ PNU

Co-Lecturer @ SNU

Teaching Assistant @ SNU

Invited Talks

Conferences & Symposia

Institutional Seminars

Beyond the
five senses.
Extended physical
intelligence.