Rong Gu, Ph.D.
Postdoc at Mälardalen University, Sweden
Office: U1-048B, Västerås Campus
Work phone: +46(0)736621483
Email: rong.gu at mdu.se
Official page: http://www.es.mdh.se/staff/3552-Rong_Gu
CV: download
Research Profile
Research Description
My research is about applying formal methods and machine learning in autonomous systems. Machine learning bears the promise of letting machines learn by themselves given enough training data. However, it has no guarantee of correctness, safety, and security, which are crucial for autonomous systems. My research aims to overcome these shortcomings via formal methods as they are based on mathematics and can provide rigorous analysis such as formal verification.
Projects as PI
SATISFIES (Holistic Synthesis and Verification for Safe and Secure Autonomous Vehicles). 2024 - 2026. KKS funded. Industrial partners: Volvo Car, Zenseact, Mimer Information Technology.
Academic Activities
Reviewer/sub-reviewer: FM Symposium, AST, NFM, FMICS, SERENE, ISEC, Journal of Robotics, IEEE Transactions on Intelligent Transportation Systems, Journal of Supercomputing, Journal of Software and Systems Modeling, Journal of Systems & Software, etc.
Organizing conferences/workshops: SSE 2024 (PC member), SEAA 2024 (PC member), ASYDE 2024 (PC member), SPIN 2024 (PC member). ECBS 2023 (Tool Demo Chair), MODELS 2023 (Volunteer), ICST 2018 (Volunteer).
Invited seminars/conferences: Dagstuhl Seminar 24071 - Safety Assurance for Autonomous Mobility. FM Symposium 2024, journal first paper. QUATIC 2023, journal first paper.
Collaborations with Industry
Volvo Car. Volvo Construction Equipment. Zenseact. Mimer Information Technology.
Teaching Profile
Teaching:
PhD course: Formal Methods (to be announced)
Teaching assistant:
Catching Bugs by Formal Verification (2018 - now)
Development of Web Application (2017 - 2019)
Data Communication (2018 - now)
Programming with Python (2022 - 2023)
Lecture:
Guest lecturer of course: Embedded Systems II. 2021 - now.
Supervision:
Supervisor of Bachelor theses.
Model Checked Reinforcement Learning For Multi-Agent Planning. 2023.
Synthesis and Verification of Neural Networks by Using UPPAAL. 2023.
Supervisor of Master theses.
Overcoming the ambiguity and inconsistency of requirements by using generative AI. 2024.
Safety-Guaranteed Mission Planner for Autonomous Vehicles. 2020.
Co-supervisor of PhD students
Selected Presentations
LiVe'24 - Integrating the Power of Machine Learning and Model Checking in Safety-Critical Systems
FM'21 - Model Checking Collision Avoidance of Nonlinear Autonomous Vehicles
FMICS 2020 - Verifiable and Scalable Mission-Plan Synthesis for Autonomous Agents