Bartocci, E. (2018). Monitoring, Learning and Control of Cyber-Physical Systems with STL (Tutorial). In C. Colombo & M. Leucker (Eds.), Runtime Verification (pp. 35–42). Springer. https://doi.org/10.1007/978-3-030-03769-7_4
Proc. of RV 2018: the 18th International Conference on Runtime Verification
10-Nov-2018 - 13-Nov-2018
Limassol, Cyprus, EU
Number of Pages:
Signal Temporal Logic (STL) is a popular specification language to reason about continuous-time trajectories of dynamical systems. STL was originally employed to specify and to monitor requirements over the temporal evolution of physical quantities and discrete states characterizing the behavior of cyber-physical systems (CPS). More recently, this formalism plays a key role in several approaches for the automatic design of safe systems and controllers satisfying an STL specification. However, requirements for CPS may include behavioral properties about the physical plant that are not always fully known a-priori and indeed cannot be completely manually specified. This has opened a new research direction on efficient methods for automatically mining and learning STL properties from measured data. In this tutorial we provide an overview of the state-of-the-art approaches available for monitoring, learning and control of CPS behaviors with STL focusing on some recent applications.
Computer Engineering and Software-Intensive Systems: 100%