Modeling Methodologies For Cyber-Physical Systems .
2 46Modeling Methodologies for Cyber-PhysicalSystems: Research Field Study on Inherent andFuture ChallengesImran Quadri, Alessandra Bagnato, Etienne Brosse and Andrey SadovykhSofteam Research & Development Division; Paris, France. Email: FirstName.LastName@softeam.frAbstractThis paper presents a field study about the inherentchallenges involved in the design of Cyber-PhysicalSystems and how Model-Based Design (MBD) iscurrently being utilized in various researchdirections in order to help in the design anddevelopment of these complex systems. The paperthen presents the first future directions andchallenges deemed to be tackled by H2020 INTOCPS project. The aim of INTO-CPS project is tocreate an integrated “tool chain” for comprehensiveModel-Based Design of Cyber-Physical Systems(CPSs). The tool chain will support themultidisciplinary, collaborative modeling of CPSsfrom requirements, through design, down tohardware and software implementation. This willenable traceability at all stages of the developmentprocess. The paper aims at analyzing the currentState-of-the-Art related to CPSs and be a basis tofuture extensions of the SysML standard to supportCPS modeling within the INTO-CPS project.Keywords: MBD, Cyber-Physical Systems, INTOCPS1 IntroductionCyber-Physical Systems can be considered as the nextgeneral of Embedded Systems. In recent years, the growthof connected Cyber-Physical Systems (CPSs) and Internetof Things (IoT) devices has increased tremendously dueto the availability of high-capacity networks (3G and4G/LTE networks), advanced sensors (e.g. RFID, NFC,etc.), protocols (e.g. IPv6, MQTT, etc.), mobile Internetand wearable devices. This paradigm shift will acceleratein coming years to drive the next technological revolutionfor CPSs, where a plethora of light-weight interconnecteddevices will be able to interact, communicate and sharevast amounts of data.Cyber-Physical Systems and especially Model-basedCPSs methodologies as an emerging area of increasingrelevance require a comprehensive framework for theirvalidation and certification. This includes both, thevalidation and certification of the embedded devices(sensors and actuators), as well as of the optional Cloudbased services which can take over the computation ofcritical aspects of the CPSs operation. This paper aims toVo lu me 3 6 , Nu mb er 4 , Dec ember 2 015provide an overview of the current research activitiesregarding Model-based methodologies for CPSs andincludes the related challenges involved in developingthese complex systems, along with the objective todetermine the current and future directions related toCPSs. The paper will also serve as a foundation for futureextensions of the SysML standard to support CPSsmodeling and to the clear semantics on SysML usage thatwill be carried out within the INTO-CPS research project.2 Cyber-Physical SystemsThe wide number of application areas of CPSs demanddesign technologies able to cover various industrialdomains like automotive, industrial control, medical,mobile communication, etc. Each domain has differentpoint of views on the underlying technical and physicaldetails. By observing the different challenges that areinherit in the design of CPSs, it is evident that CPSs needimproved multi-disciplinary modeling and specificationmethodologies able to support static analysis, verification,simulation, performance analysis and implementationtechnologies [1]. To address these issues, domain specificlanguages have been developed to cover the designchallenges of specific design domains. For example, afamous example can be found in the automotive domain:AUTOSAR (AUTomotive Open System ARchitecture) isthe de-facto standard for automotive software and E/E(Electrics/Electronics) architectures [2]. It provides abasic infrastructure to assist with developing vehicularsoftware using Atomic Software Components, running ona standardized middleware layer, called Run-TimeEnvironment (RTE).Furthermore, it includes the standardization of basicsystems functions, enables scalability to different vehicleand platform variants and upgrades over the vehicle'slifetime. Various commercial AUTOSAR systemdevelopment tools are available on the market: dSpaceSystemDesk, ETAS ISOLAR-A, KPIT K-SAR, MentorGraphics Volcano Vehicle Systems Architect, and VectorInformatik PREEVision [3] and DaVinci Developer,targeting the automotive domain specific solutions.Standards such ANSI/ISA-5.1-1984 have been used tospecify CPSs by making use of process models todescribe measuring and control devices. While theseprocess models are able to describe different properties ofAd a User Jo urn al
I. Q uadr i, A. Bag na to, E. Brosse a nd A. Sad o vykhthe physical environment, they cannot adequately coverthe computational architecture details.2.1 Modeling of Cyber-Physical SystemsModel-Based Design (MBD) has been identified as apowerful design technique for CPSs [4]. In MBD, modelsare at heart of the design process. Specifications of systemand its underlying components are defined in the form ofmodels able to reflect the evolution of the system. Thesemodels can be used for early design analysis; can help inseparation of concerns, traceability, trace generation,impact analysis, formal verification, simulation andsynthesis. By making use of models, it is possible to haveearlier identification of design defects instead of duringthe prototyping phase at a much higher cost. Additionally,automated or semi-automated processes can also help tosynthesize implementations from models, such asautomatic code generation and software synthesis onheterogeneous platforms [5]. However, the intrinsicheterogeneity and complexity of CPSs stresses all existingmodeling languages and frameworks, and, currently, it isnot possible for a single modeling language or tool toadequately address all challenges related to CPSs.For CPSs modeling, a large number of modelinglanguages have been utilized to address the underlyingaspects such as physical processes and requirementsmanagement. A good survey has been made coveringlanguages and tools like Stateflow/Simulink, Modelica,Checkmate and Massaccio; by the Columbus project [6]in order to define an interchange format for CPSs. Theselanguages enable CPSs modeling for design phases suchas simulation and verification. In [7], the authorsintroduce a test bed for collaborative control andinformation acquisition for maritime applications. Anabstraction language in the form of a Domain SpecificLanguage (DSL) for implementation of mission-levelcontrollers has been developed, termed as theCollaborative Sensing Language (CSL).Recently, high level languages such as UML [8], SysML[9] and MARTE [10] have also been utilized for modelingthese complex distributed systems. However, as statedbefore, none of these languages can singly address all thechallenges related to CPSs modeling. UML, traditionallyused for modeling of software systems, defines the syntaxof model diagrams; it does not offer any specificsemantics for CPS modeling. The OMG SysML standarddoes offer aspects such as requirements managementwhich can be interesting for CPSs, but suffers fromhaving many semantic alternatives, which are usuallyprovided by tool vendors [11] and does not providemanners to define characteristics of real-time embeddedsystems such as non-functional constraints, and aspectslike performance and energy consumption. MARTE,which is the recent OMG standard for real-time embeddedsystems does enable designers to define non-functionalconstraints, but suffers from the same pitfall of not havingdetailed guidelines and semantics, which can be used bysystem designers. In absence of concrete MARTE usageguidelines, designers can be plagued with the problem ofAd a User Jo urn al2 47correct utilization of the profile concepts, for exampleincorrect utilization of a MARTE hardware processorstereotype on a port, which while is possible in thestandard, is not logical from the current hardware designpoint of view.Additionally, high level languages have also been usedfor modeling aspects of Cloud computing, software testsor services resulting in development of CloudML [12],UTP [13] and SoaML respectively [14]. UTP can beefficiently applied to foster early testing [15] and toestablish test automation by generation of executable testscripts [16]. UTP has been applied to various industrialand research case studies to increase automation in testexecution and test design in various domains, such astelecommunications, enterprise services choreographiesand eHealth [17].In short, many modeling languages have been used todescribe CPSs or aspects which can be used in CPSsdevelopment. However, modeling techniques that addressonly the software aspects are not able to accuratelyspecify CPSs. The complexity of CPSs design demandsthe usage of new system models/analytical tools, andsoftware simulation tools, along with modeling languagesand appropriate learning mechanisms [18, 19] that areable to take into account aspects related to the physicalprocesses of CPSs. In the modeling process, systems areusually considered as static entities where current orgeneral system characteristics are used to emulatesystem’s behavior. Thus for the modeling of CPSs,effective semantics are still needed able to integrate anylanguage to reap the benefit offered by MBD. Extendingboth SysML for CPSs modeling and MARTE for theunderlying embedded devices, while integrating bothmodeling languages under a common, homogeneousframework able to support a holistic modeling of complexheterogeneous CPSs, is still an open problem that needs tobe addressed.2.2 Projects carrying out CPS researchWe now look at some of the research related to CPSs inrecent years.The CHESS project [20] focuses on improving Modelbased Design (MBD) practices and technologies to betteraddress issues such as safety, reliability, performance,robustness and other extra-functional concerns for realtime and dependable embedded systems. The projectaddresses the challenges related to compositionalstructure, interactions and behavior of system componentswhile guaranteeing their correctness and the level ofservice at run time. The tool set developed in the projectsupports verification of extra-functional properties ofdifferent system components. The project also proposes amulti-concern component modeling language and editorto fits multiple industrial domains. The language proposedin the project extends UML, SysML and MARTElanguages.In the CONTREX project [21], a UML/MARTEmethodology for distributed, mixed-critical embeddedVo lu me 3 6 , Nu mb er 4 , Dec ember 2 015
2 48M od el in g M e th odo log ies f or C yb er - Ph ys ica l S ys te ms : I n here n t a nd Fu tur e C h al le ng essystems has been proposed. This modeling effort isfocused on extending the standards to integrate aspectsrelated to distributed networks and mixed- criticalitysystems, which are not fully addressed in the standards.While CPSs modeling is not undertaken in the project, theoutputs of this project can serve as a foundation for newresearch activities related to CPSs oriented Model-Basedmethodologies in the near future.The CONCERTO project [22] proposes a methodologyfor enabling correct-by-construct component assembly formulticore systems. The automatic generation of virtualprototypes has been made possible, along withintroduction of support for separation of concerns using ameta-model based approach. New run-time monitoringmechanisms have been developed to analysis extrafunctional properties such as energy consumption.Finally, the project enables iterative development byenabling back propagation from platform-specific toplatform-independent models.The COMPASS project provides tools and techniques tosupport a model-based approach to developing Systems ofCPSs, also called Systems of Systems (SoSs) byintroducing the COMPASS Modeling Language (CML)[23]. They extend SysML by the addition of formal CMLnotations. COMPASS augments CPSs modeling bymeans of additional tools and techniques to enableinformal SoS development to be undertaken under theguidance of CML analysis techniques, some of which canbe presented at the SysML level. The DESTECS project[24] proposes a methodology for defining co-modelsallowing discrete event (DE) and continuous time (CT)models to be co-simulated. The DE and CT models arelinked through a common interface specification thatidentifies shared (monitored/controlled) variables, designparameters and events. While the project supports cosimulation, verification is not supported.2.3 Recent methodologies for Cloud ComputingThere have also been interesting design methodologieswhich could help in future CPSs integrating cloudcomputing aspects.The DreamCloud project [25] aims to enable dynamicresource allocation in many-core embedded and highperformance systems, while ensuring appropriateguarantees on non-functional properties such asperformance and energy efficiency. The project focuseson integrating embedded systems with cloud-likecapabilities, in order to allow the systems to tune resourceusage in a dynamic manner without sacrificing nonfunctional constraints. The MODAClouds project [26]aims to develop novel methods along with an open-sourceand IDE run-time environment for the high-level design,early prototyping, semi-automatic code generation andautomatic deployment of applications on multi-Cloudswith guaranteed QoS. The project aims to use Modelbased Design, coupled with risk analysis to monitorapplications at run-time and optimize them based on thereceived feedback.Vo lu me 3 6 , Nu mb er 4 , Dec ember 2 0152.4 Modeling of Cloud-based CPSsCloud-based CPSs can be seen as an upcoming newdomain that aims to integrate the paradigms of Cloudcomputing and apply it to CPSs. This merger of the twoexisting domains has been termed as cyber-physicalCloud computing (CPCC) [27, 28]. Examples of CCPCinvolve utilization of smart phones in cloud sourcing,Unmanned aerial vehicles (UAVs) equipped withcameras/sensors for gathering data [29], and in-vehiclesmart phones serving as internet-connected sensors formonitoring road conditions [30]. In [31], the authorspropose a novel service provisioning model in order tomake CPCC possible as a service, while in [32], a test bedfor CPCC is developed (for simulation and actualprototype testing) that includes a diverse range of hosts:such as automobiles (cars, buses, etc.); people with smartphones; and unmanned aerial vehicles (UAVs).Additionally, various factors can be seen as challenges toCPCCs. Traditional virtualization technology, one of thekey underlying technologies for enabling secure andscalable Cloud computing, typically implementsmachine/process models which are only deterministic innature. Thus, for the same input, repeated execution ofsequential programs in these models result in the sameoutput, but other non-functional properties such as energyconsumption may vary. Therefore, virtualizationtechnology for CPCC must develop solutions that providedeterministic behavior for non-functional properties suchas execution time. Factors such as number and location ofserver machines, resource allocation, effective schedulingand management can also hinder the development ofeffective CPCCs [33].2.5 Model-Based RepositoriesIn [34] the authors presented a survey of several industrialprojects where research activities are being carried outrelated to Model-Based Repository Engineering (MBRE),and it was observed that there is currently nomethodological and unified framework for MBRE, whichin fact can benefit from well-defined and widely appliedModel-based technologies: wide range of modelinglanguages, transformation techniques, and opensource/commercial tools. The MBRE can be effectivelyused in Model-based CPSs methodologies, where largenumber of system components need to be handled andmay require re-use and deployment.In the MADES project [35], the authors develop aModel-Driven methodology for embedded systems thatdeveloped a component based model repository with thegoal of promoting component reuse in the MADESenvironment for avionics domain. The repository containsknowledge gathered about components during the projectdesign, implementation, validation and verificationphases, as well as the information about the variouscomponent constraints. The repository also providesquerying and persistence of previously available contents.The RAS (OMG Reusable Assets Specification)Container has been used in the project [36].Ad a User Jo urn al
I. Q uadr i, A. Bag na to, E. Brosse a nd A. Sad o vykhModeling tools have also started integrating MBRE intheir environments as well. The Modelio environmentnow offers the notions of worldwide modeling [37] whererepositories are managed by means of model fragments,which can be distributed and shared between users, andimpact analysis and consistency management can becarried out across the entire scope of a system. In [38] theauthors illustrate the research results of collaboration withCisco to increase the cost-effectiveness of black-boxsystem-level testing for Video Conferencing Systems(VCSs). For this purpose, test case repositories have beendeveloped and automatically generated test cases resultingfrom the research output are stored separately in two otherseparate repositories. In addition, two UML repositoriesat the product line and product levels have been built forgenerating test cases. Similarly in [39], researchcollaboration with FMC Technologies, a leading globalprovider for technology solutions for energy industry hasresulted in development of two repositories to manage therequirements at both the product line and product levels.Requirements are specified using a restricted use casemodeling approach, which are transformed into restrictedtextual test case specifications automatically.2.6 Virtualization of CPSsThe current state of the virtualization technologies is theresult of a convergence of several areas includingprocessor architectures, operating systems, compilers andcommunication networks. Hypervisors are layers ofsoftware that exploit the features of the hardware platformto establish independent execution environments. Severalvirtualization solutions can be distinguished such as fullvirtualization, OS virtualization support and bare metalhypervisors.Several projects are nowadays addressing multicorevirtualization where time and space isolation of hardwareis critical for engineering systems with mixed criticalities(e.g. safety or security). MultiPARTES [40] project isaddressing multicore partitioning and introducing a novelModel-Driven approach for virtualization based onmulticore partitioning. DREAMS [41] project isgeneralizing this approach by introducing an architecturalstyle to broaden adoption to further domains. In a similarvein, PROXIMA [42] project is addressing virtualizationfrom a probabilistic perspective. The focus ofvirtualization has shift recently from low level hardwareresources such as memory and processors tocommunications (e.g. shared buses, networks-on-chips,etc.). The Cloud poses the next frontier to virtualization inCPSs.2.7 Programming Cloud-based CPSs service unitsSeveral types of software service units, such as forInfrastructure-as-a-Service (IaaS), Platform-as-a-Service(PaaS), Software-as-a-Service (SaaS), and Data-as-aService (DaaS), have been available and widely used inpractice. Similarly, several works have been developedfor integrating IoTs/CPSs into cloud system such as in[43]. From the programming perspective, someframeworks have been developed to simplify theAd a User Jo urn al2 49programming of cloud services, such as JClouds, Boto,Aneka, BOOM and OpenStack. They abstract cloudresources and support different programming models,such as MapReduce and dataflows. However, to datethere is a lack of programming models/frameworks thatprovide a unified way and programmable APIs forexecuting applications in Cloud-based CPSs. Severalworks have presented dynamic programming capabilitiesof
future extensions of the SysML standard to support CPS modeling within the INTO-CPS project. Keywords: MBD, Cyber-Physical Systems, INTO-CPS 1 Introduction Cyber-Physical Systems can be considered as the next general of Embedded Systems. In recent years, the growth of connected Cyber-Physical Systems (CPSs) and Internet
Bruksanvisning för bilstereo . Bruksanvisning for bilstereo . Instrukcja obsługi samochodowego odtwarzacza stereo . Operating Instructions for Car Stereo . 610-104 . SV . Bruksanvisning i original
Cyber Vigilance Cyber Security Cyber Strategy Foreword Next Three fundamental drivers that drive growth and create cyber risks: Managing cyber risk to grow and protect business value The Deloitte CSF is a business-driven, threat-based approach to conducting cyber assessments based on an organization's specific business, threats, and capabilities.
10 tips och tricks för att lyckas med ert sap-projekt 20 SAPSANYTT 2/2015 De flesta projektledare känner säkert till Cobb’s paradox. Martin Cobb verkade som CIO för sekretariatet för Treasury Board of Canada 1995 då han ställde frågan
service i Norge och Finland drivs inom ramen för ett enskilt företag (NRK. 1 och Yleisradio), fin ns det i Sverige tre: Ett för tv (Sveriges Television , SVT ), ett för radio (Sveriges Radio , SR ) och ett för utbildnings program (Sveriges Utbildningsradio, UR, vilket till följd av sin begränsade storlek inte återfinns bland de 25 största
Hotell För hotell anges de tre klasserna A/B, C och D. Det betyder att den "normala" standarden C är acceptabel men att motiven för en högre standard är starka. Ljudklass C motsvarar de tidigare normkraven för hotell, ljudklass A/B motsvarar kraven för moderna hotell med hög standard och ljudklass D kan användas vid
LÄS NOGGRANT FÖLJANDE VILLKOR FÖR APPLE DEVELOPER PROGRAM LICENCE . Apple Developer Program License Agreement Syfte Du vill använda Apple-mjukvara (enligt definitionen nedan) för att utveckla en eller flera Applikationer (enligt definitionen nedan) för Apple-märkta produkter. . Applikationer som utvecklas för iOS-produkter, Apple .
ing. Modern power systems are thus cyber-physical power systems (CPPS). Although the coupling of these two net-works brings some convenience, the power system is more vulnerable to intricate cyber environment, which puts the CPPS at the risk of cyber attacks [1], [2]. In general, external attacks on CPPS can be divided into physical attacks, cyber
1 Archaeological Laboratory Techniques [8/2015]. Suggested Reading. Adkins, Lesley, and Roy Adkins . 2009 . Archaeological Illustration. Paperback ed. Cambridge Manuals in
