SysML for Modeling Co-Simulation Orchestration over FMI, INTO-CPS Approach
1 like1,024 views
1) The document discusses SysML modeling of co-simulation orchestration over the Functional Mockup Interface (FMI) as part of the INTO-CPS project. 2) The INTO-CPS project aims to build an integrated model-based design toolchain for cyber-physical systems covering the full development lifecycle. 3) Modelio is a UML modeling tool that provides SysML diagrams and modules for modeling CPS within the INTO-CPS framework.
![alessandra.bagnato@softeam.fr 18
System Interface Modelling
• Internal Block Diagram
– Divide into CT/DE constituent models/systems/components
– Define interfaces between different components
ibd Robot
«block»
Controller
«block»
MotorLeft
«block»
WheelLeft
«block»
EncoderLeft
«block»
SensorLeft
«block»
MotorRight
«block»
WheelRight
«block»
EncoderRight
«block»
SensorRight
«block»
Body
motorSignalL:real
encoderSignalL:real encoderSignalR:real
motorSignalR:real
rotationL:rad/s rotationR:rad/s
toWhellRtoWheelL toBodyRtoBodyL
sensorSignalR:realsensorSignalL:real
ibd Environment
«block»
Line
robotPosition:real[2,1] opticalReflectionR:realopticalReflectionL:real](https://image.slidesharecdn.com/decps2016-into-cpsv5-presentation-160622083935/85/SysML-for-Modeling-Co-Simulation-Orchestration-over-FMI-INTO-CPS-Approach-18-320.jpg)
SysML for Modeling Co-Simulation Orchestration over FMI, INTO-CPS Approach
- 1. www.into-cps.au.dk SysML for Modeling Co-Simulation Orchestration over FMI, INTO-CPS Approach Alessandra Bagnato SOFTEAM R&D Workshop Challenges and New Approaches for Dependable and Cyber-Physical System Engineering (De-CPS 2016) Pisa, 17th June 2016 21st International Conference on Reliable Software Technologies Ada-Europe 2016
- 2. SOFTEAM Cadextan / IDENTITY IT Consulting 10% Business Consulting 25% Technolog y Consulting 65% Capital 20 M€ Sales 90 M€ Staff 930 employees Investment Banking, Asset Management 50% Energy, Services 20% Retail banking 20% Insurance 10% LONDRES SINGAPOUR Toulouse TUNIS Senior engineers 650 Architects 250 .Net, JEE, C++, web, BI Architects, urbanists, experts, project managers, project directors Confidential, SOFTEAM Cadextan 2
- 3. alessandra.bagnato@softeam.fr • UML editor with 25 years’ history – SysML, MARTE, BPMN – Code generation – Documentation – Available under open source at Modelio.org 3 Modelio for System Engineering www.modelio.org Modelio 3
- 4. alessandra.bagnato@softeam.fr • Modelling with UML, SysML and BPMN • Requirements Modeling • Tabular editors • Import/export MS Excel & Word • Embedded Systems modelling via MARTE • Traceability Editor • Impact Analysis • Document Generation • Support for Collaborative activities (Constellation, SVN) • Automatic diagrams creation • Customisable, interfaces to external tools 4 Modelio System Architect Solution Dedicated to System architects modelling with SysML, UML or BPMN and carrying out Requirements based analysis Traceability Editor Dedicated SysML editor
- 5. alessandra.bagnato@softeam.fr 5 Cyber-Physical Challenges Agricultural Logistics Smarter Building Design • Models of large-scale, open, diverse data integration • … coupled with models of physics • Need to model learning behaviour • Possible integration of models of human behaviour • Need to model control and planning/re-planning • Models of locality and mobility • Real-time behaviour modelling • Domain-specific reference models
- 6. www.into-cps.org INTO-CPS: a 8 M€ H2020 Project An integrated “tool chain” for comprehensive Model-Based Design of Cyber-Physical Systems 6
- 7. alessandra.bagnato@softeam.fr 1. Build an open, well-founded tool chain for multidisciplinary model-based design of CPS that covers the full development life cycle of CPS 2. Provide a sound semantic basis for the tool chain 3. Provide practical methods in the form of guidelines and patterns that support the tool chain 4. Demonstrate in an industrial setting the effectiveness of the methods and tools in a variety of application domains. 5. Form an INTO-CPS Association to ensure that project results extend beyond the life of the project 7 INTO CPS Objectives
- 9. alessandra.bagnato@softeam.fr 9 CPS co-modelling shared network shared computing co-simulation (MiL) environment model models of cyber elements models of physical elements real code real plant code generation realisation laboratory testing SiL HiL co-model architecture models stub model generation requirements test automation automated co-model analysis design space exploration model checking analysisplug-ins Andrey.Sadovykh@softeam.fr
- 12. alessandra.bagnato@softeam.fr • Modern systems are complex • To cope with this, we can build models beforehand – To perform analysis (e.g. static analysis, proof, model checking, simulation) – Clarify our assumptions – Evaluate potential designs – Avoid expensive prototypes • Different modelling paradigms for different aspects 12 Model-driven Design
- 13. alessandra.bagnato@softeam.fr 13 The Initial INTO-CPS Vision
- 14. alessandra.bagnato@softeam.fr • SysML – Use Case diagrams – Requirements diagrams • Informal (link and traceable) • Formal (LTL, Test automation) 14 Requirements Modelling uc Line following robot Line following Robot Controller Line Motors Sensors Read optical sensor values Adjust motor control signals Measure optical reflection «include» Encoders Wheels Read encoder ticks Measure encoder ticks «include» req FaultModelling Model faulty behaviour in the sensors Id=s1 «requirement» faultyBehaviour Model ambient light as noise in the optical sensors Id=s1.1 «requirement» ambientLight Model AD conversion erros in the LSB’s of the optical sensor readings Id=s1.2 «requirement» conversionError Model malfunctioning sensors that continuously reads the same value Id=s1.3 «requirement» sensorMalfunction Model ambient light as noise in the optical sensors Id=s1.1 «requirement» ambientLight «block» SensorRight «satisfy»
- 15. alessandra.bagnato@softeam.fr 15 The Initial INTO-CPS Vision
- 16. alessandra.bagnato@softeam.fr • The Functional Mockup Interface (FMI) is a tool- independent standard to support both model exchange and co-simulation of dynamic models using a combination of XML-files and compiled C-code • The FMI standards currently specify two types of protocols: – FMI for Model Exchange (import and export), and – FMI for Co-Simulation (master and slave). • For FMI Model Exchange Import, the subsystem model is exported from a simulation tool in the form of an FMU archive containing the necessary FMU information (model description file, optional C source code, etc.); while in the FMI Model Exchange Export, the subsystem model is imported into the simulation system for system simulation. 16 Functional Mockup Interface (FMI)
- 17. alessandra.bagnato@softeam.fr • Block Definition Diagram (top level) 17 System Decomposition bdd Line following robot «block» Robot «block» Controller «block» SensorLeft «block» SensorRight «block» WheelLeft «block» WheelRight «block» LineFollowingRobot «block» Environment «block» Line «block» Body «block» EncoderLeft «block» EncoderRight «block» MotorLeft «block» MotorRight
- 18. alessandra.bagnato@softeam.fr 18 System Interface Modelling • Internal Block Diagram – Divide into CT/DE constituent models/systems/components – Define interfaces between different components ibd Robot «block» Controller «block» MotorLeft «block» WheelLeft «block» EncoderLeft «block» SensorLeft «block» MotorRight «block» WheelRight «block» EncoderRight «block» SensorRight «block» Body motorSignalL:real encoderSignalL:real encoderSignalR:real motorSignalR:real rotationL:rad/s rotationR:rad/s toWhellRtoWheelL toBodyRtoBodyL sensorSignalR:realsensorSignalL:real ibd Environment «block» Line robotPosition:real[2,1] opticalReflectionR:realopticalReflectionL:real
- 19. alessandra.bagnato@softeam.fr • Parametric Diagram – Define continuous behaviour of CT components • State Machines (DE models generated for tests) – Define discrete behaviour of DE components 19 System Behaviour par Robot totalForce:N forceL forceR forceR:N forceR forceL:N forceL genForceL rxForceL rxForceR genForceR rxForceL:N rxForceL forceR rxForceR:N rxForceR forceL genForceL genForceR bdd Constraints «block» Robot parameters totalForce:N forceL:N forceR:N constraints {totalForce=forceL+forceR} «constraint» TotalForce parameters forceL:N genForceL:N rxForceL:N constraints {forceL=genForceL+rxForceL} «constraint» ForceLeft parameters forceR:N genForceR:N rxForceR:N constraints {forceR=genForceR+rxForceR} «constraint» ForceRight parameters rxForceL:N forceR:N constraints {rxForceL=forceR} «constraint» RxForceLeft parameters rxForceR:N forceL:N constraints {rxForceR=forceL} «constraint» RxForceRight
- 20. alessandra.bagnato@softeam.fr 20 INTO-CPS Diagrams and INTO-CPS Blocks
- 21. alessandra.bagnato@softeam.fr 21 Modelio for INTO-CPS (1/3) Architecture Diagram
- 22. 22 Modelio for INTO-CPS (2/3) Connection Diagram
- 23. 23 Modelio for INTO-CPS (3/3) Modelio INTO-CPS module
- 24. alessandra.bagnato@softeam.fr • AGCO, Denmark • Alcatel-Lucent, Ireland • Almende, Netherlands • Altran, UK • Bachmann electronic, Netherlands • Bakker Sliedrecht Electro Industrie, Netherlands • Carrier, France • CeTIM, Netherlands • Chemring TS, UK • Conpleks Innovation, Denmark • Dredging International, Belgium • DSTL, UK • Goodrich, UK • Grundfos, Denmark 24 Initial Industrial Follower Group • GN Resound, Denmark • HMF, Denmark • Huisman Equipment, Netherlands • Irmato Industrial Solutions, Netherlands • Jaguar Land Rover, UK • National Institute of Informatics, Japan • ONERA, France • Rockwell-Collins, France • Rolls-Royce, UK • Seluxit, Denmark • Siemens, Sweden • Terma, Denmark: • Thales, France • UTC Aerospace Systems, UK • West Consulting, Netherlands
- 25. Contacts Alessandra Bagnato, SOFTEAM Alessandra.bagnato@softeam.fr Modelio Web Site : http://www.modelio.org http://forge.modelio.org/projects/intocps INTO-CPS Web Site http://into-cps.au.dk/ @https://twitter.com/IntoCps https://github.com/into-cps 25 IT Services / UML Tools Research Management Standardization MONDO Project, a scalable modeling and model management on the Cloud