Welcome to Intelligent Design, Optimization, and Learning (IDOL) Lab!

Design is a universal activity involving the creation of anything, and we envision innovating design by integrating human knowledge and artificial intelligence.

Contact Us
Kang, SungKu

Kang, SungKu (CV)


Contacts

  • Email: kang.sungku@kau.ac.kr
  • Phone: 02-300-0176
  • Office: Room 403, Mechanical Engineering Building, Korea Aerospace University, Goyang

Kang, SungKu is an assistant professor in the Department of Mechanical and Aircraft System Engineering at Korea Aerospace University. He received his B.S. degree in Mechanical and Aerospace Engineering from Seoul National University in 2012, and his Ph.D. in Mechanical Engineering from the University of Illinois at Urbana-Champaign in 2017.

His research interest spans both the foundations in the extraction of human domain knowledge in a computer-understandable form, and its application to design automation through the integration with artificial intelligence, especially reinforcement learning.

Prior to joining Korea Aerospace University, he worked at Northeastern University and Virginia Tech, where he contributed to research projects in design automation for sustainable building (e.g., stone masonry), renewable energy generation (e.g., offshore wind turbine), and personalized products. In his spare time, he likes cooking with his wife.

Professional Experience

  • Korea Aerospace University, Department of Mechanical and Aircraft System Engineering, Assistant Professor (2024-)
  • Northeastern University, Department of Civil and Environmental Engineering Department, Postdoctoral Research Fellow (2020-2024)
  • Virginia Polytechnic Institute and State University, Grado Department of Industrial and Systems Engineering, Postdoctoral Research Associate (2018-2020)
  • General Electric Global Research Center, Graduate Research Internship (Summer 2015)

Education

  • Ph.D., Mechanical Engineering, University of Illinois at Urbana-Champaign, 2017
  • B.S., Mechanical and Aerospace Engineering, Seoul National University, 2012

Teaching

  • Computer-Aided Design (AM3211)
  • Manufacturing Processes for Engineering Materials (AM4335)
  • Mechanical Engineering Design (AM4321)

Research Topics

Integration of Domain Knowledge with Artificial Intelligence

Artificial intelligence is being adopted for important tasks in design and manufacturing, but not yet deeply integrated with the domain knowledge to guide the holistic design process. This is because a considerable amount of domain knowledge is maintained in a format not readily understandable by a computer. To integrate insightful domain knowledge with artificial intelligence for design automation, this research aims at utilizing natural language processing and generative models to extract domain knowledge in a computer-understandable form.



Holistic Design Automation via Reinforcement Learning

Currently, supply chain, product, and controller design involve separate design iterations rather than a holistic design paradigm, thus limiting the efficiency of the entire design process. Founded upon reinforcement learning, Google AlphaStar mastered the real-time strategy game StarCraft II, and demonstrated a holistic model that enables prompt and flexible decision-making in resource harvesting, unit production, and real-time strategic control, where each of the tasks is analogous to supply chain, product, and controller design respectively. Inspired by the success of reinforcement learning, the objective of this research is to realize holistic design of supply chain, product, and controller through reinforcement learning.



Design Automation for Sustainablility

Stone masonry with locally sourced natural stones is considered as a sustainable form of construction. However, its adoption is limited, as it is challenging to determine a proper sequence of natural stones with random sizes/shapes in real time. Inspired by the analogy between stone masonry and the classic game of Tetris, Tetris AI is customized to autonomously learn stone masonry design in a virtual environment by playing Tetris-like games.

Team

Become a member of IDOL Lab!

Output

Projects

Research papers

Conferences

Other Publications