Intelligent Test Automation and Optimization
Reinforcement learning agents play a game themselves to find defects, leaving bug reports and input-sequence logs behind. Rewarding error detection is what steers the search. We also predict the causes of novice programmers' errors from learning data, and grade practice code automatically by running it in a simulator.
Read more
Formal Methods and Embedded System Verification
We move requirements and designs into Petri nets and LTS to analyse reachability and safety formally. Real code is verified against behaviour models recovered from usage logs and checked by combinatorial analysis; the target is the indirect dependencies that sensor interactions create. AUTOSAR components are tested without hardware through SiL simulation.
Read more
UML-Based System Modeling and Static Verification
We extract class models from Korean requirement sentences and generate framework-independent code from state machine diagrams. Where a model and its implementation have drifted apart, and where an Android manifest is misconfigured, we find it by static analysis.
Read more
Digital Twin Simulation and Autonomous System Verification
We build virtual verification platforms that join data, AI, and CAE. For textile materials and composite structures, manufacturing data is connected to AI models, and the verification procedure is standardised so design decisions can be made before a physical prototype exists. Integrated simulation of distributed embedded systems and flight testing of unmanned aerial vehicles are part of the same line.
Read moreOngoing Research Projects
We work closely with industry on problems that have to hold up in production. These are the projects currently driving that work.
Selected Publications
Development of an automatic class diagram generator using an AI-based GRU classification model and 5W1H heuristic rulesInternational Journal
Journal of Systems and Software, Vol. 235, pp. 112780, 2026-5
Integration and analysis of use cases using modular Petri nets in requirements engineeringInternational Journal
IEEE Transactions on Software Engineering, Vol. 24, No. 12, pp. 1115-1130, 1998
An empirical study of configuration changes and adoption in Android appsInternational Journal
Journal of Systems and Software, Vol. 156, pp. 164-180, 2019-10
Developer Mistakes in Writing Android Manifests: An Empirical Study of Configuration ErrorsInternational Conference
2017 IEEE/ACM 14th International Conference on Mining Software Repositories (MSR), pp. 25-36, 2017-5
시뮬레이터를 이용한 임베디드 소프트웨어 자동 채점 시스템 및 방법Patent Granted
Granted 10-2373133 (2022-03-14)
단위 테스트 케이스 재사용 기반의 함수 테스트 장치 및 그 함수 테스트 방법 (FUNCTION TEST APPARATUS BASED ON UNIT TEST CASES REUSING AND FUNCTION TEST METHOD THEREOF)Patent Filed
Filed 14/123,297 (2012-12-18), US
Our Team
We build the future of software with the best people we can find.