Ádám, Z., López Miguel, I. D., Mavridou, A., Pressburger, T., Bęś, M., Blanco Viñuela, E., Katis, A., Tournier, J.-C., Trinh, K. V., & Fernández Adiego, B. (2023). Automated Verification of Programmable Logic Controller Programs Against Structured Natural Language Requirements (NASA/TM–20230003752). National Aeronautics and Space Administration. https://doi.org/10.34726/5291
PLCverif is an actively developed project at CERN, enabling the formal verification of Programmable Logic Controller (PLC) programs in critical systems. In this paper, we present our work on improving the formal requirements specification experience in PLCverif through the use of natural language. To this end, we integrate NASA’s FRET, a formal requirement elicitation and authoring tool, into PLCverif. FRET is used to specify formal requirements in structured natural language, which automatically translates into temporal logic formulae. FRET’s output is then directly used by PLCverif for verification purposes. We discuss practical challenges that PLCverif users face when authoring requirements and the FRET features that help alleviate these problems. We present the new requirement formalization workflow and report our experience using it on two critical CERN case studies.
en
Research Areas:
Logic and Computation: 20% Information Systems Engineering: 40% Automation and Robotics: 40%
CC BY 4.0
