Title: | Investigating reliable Co-Simulation with the FMI standard |
Subject: | Computer science, Software engineering, Embedded systems, Robotics, Dependable Aeronautics and aerospace |
Level: | Advanced |
Description: |
Background Co-simulation is a growing paradigm, which can be defined as: Joint simulation of loosely coupled stand-alone sub-simulators. The Functional Mockup Interface (FMI) standard is being continuously adopted by industry. At a glance, the standard enables simulation models to be exported to tool-agnostic black boxes, which can be integrated to create larger simulation models consisting of several models, from different originating tools or language (e.g, a Simulink model and a block of C-code). The black box FMI-aligned models contain inputs, outputs, and potentially some variable parameters (e.g, mass of a system). The FMI standard as such enables Co-simulation (among other things) through various integration mechanisms. The standard has a lot of potential for re-use of heterogenous models that otherwise would be left behind due to tool-integration issues and can enable collaboration across engineering teams. Tasks The student(s) would investigate the practical implementation of the standard to improve the useability regarding simulation accuracy and/or performance. Previous work has identified several issues that need to be addressed for the standard to reach a maturity suitable for industrial use. Some actions the student(s) could look into are: -How to achieve good accuracy of results with FMI simulation -How to improve execution time with regard to overhead -What trade-offs exist with FMI-based Co-simulation compared to traditional simulation -Suggest workflows for how to parameterize models efficiently |
Start date: | |
End date: | |
Prerequisites: |
Knowledge about simulation and tools like Simulink/OpenModelica are useful for this thesis. |
IDT supervisors: | Johan Cederbladh |
Examiner: | |
Comments: | |
Company contact: |