Principle Investigator
Soomok Lee, Ph.D (이수목 교수)
Vice Dean of the Department of AI Mobility Engineering
Vice Director of the Department of Innovative Convergence (Future car division)
Assistant Professor at the Department of Artificial Intelligence
I recently joined as an assistant professor in the Department of AI Mobility Engineering and Mechanical Engineering at Ajou University. My research interest covers artificial intelligence, 2D/3D computer vision, machine learning, sensor fusion and localization in various autonomous mobility platforms. Before joining Ajou University, I performed a machine learning-oriented postdoc researcher at the center of advanced computing and analysis at Harvard University (HMS/MGH), Boston MA. My research interest lies in 2D/3D object detection and tracking, federated learning, continual learning, domain adaptation, and multi-modal fusion. I also performed as a Director of Software Engineering of ThorDrive US, located in Cincinnati OH, a venture-backed startup creating ground-up, fully autonomous vehicles for point-to-point mobility in airport ramps having project partnerships with Delta affiliates (Unifi), UPS and DHL. Prior to ThorDrive, I received the B.S. degree in electrical engineering from Sungkyunkwan University and then completed a computer science Ph.D degree at Seoul National University (SNU), Korea. There, I developed algorithms for SNU’s $100K-winning entry in the 2013 Korean Autonomous Challenge, and went on to lead a lot of projects developing not only self-driving modules with LG and Mando, but also the SNU self-driving platform, which was the first urban driving platform in Korea.
The previous teams Soomok Lee was with.
Working Experience
Assistant Professor @ AI Mobility Engineering(Undergrad.) Artificial Intelligence(Grad.) 2023 – present
Ajou University, Suwon, Korea
Assistant Professor @ Mechanical Engineering 2022 – 2023
Ajou University, Suwon, Korea
PostDoc. Researcher, 2022
Harvard University (MGH/HMS), Boston MA, United States
Director of Software Engineering, 2017 – 2022
ThorDrive, Inc., Cincinnati OH, United States
PhD candidate researcher, 2013 – 2017
SNU Intelligent Vehicle IT Center, Seoul, Korea
Visiting student under CARS visiting professor, 2014
Stanford University, Palo Alto CA, United States
Research Intern, 2012
Hyundai Power-Train Research Center, Hwaseong, Korea
Visiting Student Intern, 2010 – 2011
University of Central Florida, Orlando FL, United States
Data Analysis Intern 2010
BrightPower, Inc., Manhattan NY, United States
Education
Ph. D. degree in Electrical Engineering and Computer Science, 2013 – 2019
Seoul National University, Seoul, Korea
Bachelor of Engineering in Electronic and Electrical Engineering, 2006 – 2013
SungKyunKwan University, Suwon, Korea
English Language Institute, 2010
Queen’s College (CUNY), Queens, NY, United States
Working experience with projects
Harvard University (MGH/HMS), Boston MA, US
2022.06 - 2022.08:
Postdoctoral researcher,
Project: GE Healthcare & Rhino Health
Software Development
- 3D Multi-modal detection on brain aneurysm detection
- Federated learning algorithm development on multi-national clinical data team
ThorDrive, Inc. Cincinnati, OH, US
2019.10 - 2022.05:
Director of Software Engineering,
Project: Self-driving Ground Service Equipment (GSE) development at Airport ramp
- Airport partners: Cincinnati/Northern Kentucky (CVG) and Dayton (DAY) airports
- Service partners: UPS, DHL, and Delta Global Services
Software Development
- Development of fusion-based object detection methods in airport ramps.
- Development of multi-modal context-based object behavior attribute network using thermal, LIDAR, and RGB cameras.
- Development of intersection, airplane stop and go decision-maker, and risk evaluator
Sensor Skills
- RGB(Idscamera, Point Grey), LIDAR (Velodyne, Hesai), Thermal Camera(FLIR), Radar(continental ARS 408-21) and IMU sensors(Xsens)
ThorDrive Korea. Seoul, Korea
2017.9- 2019.10:
Team lead / Senior Perception Engineer,
Self-driving delivery service projects: Developed various self-driving delivery service platforms
- Project Partners: Ace Hardware, an US hardware retailer in CA and Emart, the largest retailer in Korea
Software Development
- Leading algorithm development including object, traffic sign and light, lane marker and parking slot detection and camera distance estimation.
- Development of multi-modal localization system using both LIDAR and cameras
- Managing two projects retrofitting Ford's product car into a self-driving platform
Seoul National University, Vehicle Intelligence Lab
Seoul, Korea
2013.01- 2019.07
Ph. D. Candidate
Project lead, Mar. 2015 – Feb. 2017
“Parking related landmark/space detection and visual localization system using AVM camera,” Funded by LG Electronics, Inc.
- Development of parking space and land-mark recognition system using AVM
- Development of map-based visual localization system for valet parking using AVM
Research Operating Manager, Aug. 2014 – Mar. 2017
Intelligent Vehicle IT center, Funded by National Research Foundation & Ministry of Science, ICT & Future planning, Korea
- Development of lane-based localization system in highly congested road environments.
- Development of road environment recognition algorithms for autonomous taxi platform
Project lead, June. 2013 – Oct. 2015
“Development of Driver Assistant System Using Camera, Radar and Road Characteristics,” Funded by MANDO Co. Ltd. & Ministry of Knowledge Economy, Korea
- Lateral Localization by camera sensor without utilization of a pre-built map
- Lane information fusion using multiple lane detection results from heterogeneous sensors
Computer skills:
OS Platforms
Programming
Engineering Tools/Libraries
Project Tools
Linux, MacOS, Windows, ROS
C/C++, Python, Matlab
PyTorch, Tensorflow, OpenCV, PCL, Simulink, MRPT, CANoe
Azure, Git, Atlassian Suite (Jira, Confluence)
Languages:
Korean Native
English Professional