A Semantic-Aware, Accurate and Efficient API for (Co-)Simulation of CPS

Giovanni Liboni (1) , Julien Deantoni (2, 3, 4)

1 Safran Tech
2 UniCA - Université Côte d'Azur
3 Laboratoire I3S - COMRED - COMmunications, Réseaux, systèmes Embarqués et Distribués
4 KAIROS - Logical Time for Formal Embedded System Design

To understand the behavior emerging from the coordination of heterogeneous simulation units, co-simulation usually relies on either a time-triggered or an event-triggered Application Programming Interface (API). It creates bias in the resulting behavior since time or event triggered API may not be appropriate to the behavioral semantics of the model inside the simulation unit. This paper presents a new semanticaware API to execute models. This API is a simple and straightforward extension of the Functional Mock-up Interface (FMI) API. It can be used to execute models in isolation, to debug them, and to co-simulate them. The new API is semantic aware in the sense that it goes beyond time/event triggered API to allow communication based on the behavioral semantics of internal models. This API is illustrated on a simple co-simulation use case with both Cyber and Physical models.

Mots clés: co-simulation; API; behavioral semantics

Domaines: Systèmes embarqués; Modélisation et simulation

Giovanni Liboni, Julien Deantoni. A Semantic-Aware, Accurate and Efficient API for (Co-)Simulation of CPS. CoSim-CPS 2020 - Software Engineering and Formal Methods. SEFM 2020 Collocated Workshops, Sep 2020, Amsterdam / Online, Netherlands. ⟨10.1007/978-3-030-67220-1_21⟩. ⟨hal-03038527⟩