Text

STGEM-System Testing Using Generative Models

Photo:https://unsplash.com/

STGEM is a tool for black-box testing of cyber-physical systems. It supports falsification of requirements described in Signal Temporal Logical (STL) by robustness optimization. This is achieved by training a generative machine learning model online to produce system inputs that yield a low robustness.

STGEM is under development and we are adding new algorithms and features. So far, the tool implements the algorithms presented in these articles:

  • Jarkko Peltomäki and Ivan Porres. Falsification of Multiple Requirements for Cyber-Physical Systems Using Online Generative Adversarial Networks and Multi-Armed Bandits. The 6th. Intl. Workshop on Testing Extra-Functional Properties and Quality Characteristics of Software Systems, ITEQS 2022.
  • Jarkko Peltomäki, Frankie Spencer and Ivan Porres. Wasserstein Generative Adversarial Networks for Online Test Generation for Cyber Physical Systems. The 15th Intl. Workshop on Search-Based Software Testing, SBST 2022.
    Jarkko Peltomäki, Frankie Spencer and Ivan Porres. WOGAN in the SBST 2022 CPS Tool Competition. The 15th Intl. Workshop on Search-Based Software Testing, SBST 2022.
  • Jesper Winsten, Ivan Porres. WOGAN at the SBFT 2023 Tool Competition - Cyber-Physical Systems Track. The 16th Intl. Workshop on Search-Based and Fuzz Testing, SBFT 2023.
  • Claudio Menghi, Paolo Arcaini, Walstan Baptista, Gidon Ernst, Georgios Fainekos, Federico Formica, Sauvik Gon, Tanmay Khandait, Atanu Kundu, Giulia Pedrielli, Jarkko Peltomäki, Ivan Porres, Rajarshi Ray, Masaki Waga and Zhenya Zhang. ARCH-COMP 2023 Category Report: Falsification, Proceedings of 10th International Workshop on Applied Verification of Continuous and Hybrid Systems (ARCH23), vol 96, pages 151-169.
  • Jesper Winsten, Valentin Soloviev, Jarkko Peltomäki, Ivan Porres, Adaptive test generation for unmanned aerial vehicles using WOGAN-UAV. Preprint available at here External link, opens in new window.