Real-time Ethernet Residual Bus Simulation:A Model-Based Testing Approach for the

Next-Generation In-Car Network

Florian Bartols Till Steinbach Franz Korf Bettina ButhThomas C. Schmidt

Hamburg University of Applied Sciencesflorian.bartols@haw-hamburg.de

October 10th, 201422nd International Conference on Real-Time Networks and Systems

RE NET

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

RESoftware and functions in modern cars

Functions are implemented mostly in software todayUtilization of software directly influences thedevelopment costsTesting in early development stages reduces thesecostsDistributed development makes early testing difficult

2 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REComplex In-Car Interconnections

On-BoardSystemsOff-BoardSystems

Engine Bus (High-Speed-CAN)

Body Bus (Low-Speed-CAN) Multimedia Bus (MOST)

Diagnostics Bus (CAN / Ethernet)

Chassis Bus (FlexRay)

Doors (LIN)

Lights (LIN)

...

...

...

...

Engine Gearbox Breakes

Door AC Head-lights

Radio/Navi

Front Chassis

Rear Chassis

Dash

boar

d CD/DVD

Cent

ral

Gat

eway

Service

The complexity of current in-car interconnections ishardly manageable

3 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REComplex In-Car Interconnections

On-BoardSystemsOff-BoardSystems

FrontChassis

RearChassis

Engine

Gearbox

Rear Breakes

Doors

ACHead-lights

Radio/ Navi

CD/DVDDash-board

Front Breakes

InternetGPS

RTEthernet Switch

RTEthernet Backbone

Service

RTEthernet Switch

RT Ethernet for in-car interconnection reduces thecomplexity

3 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REContribution

Testing systems and applications in earlydevelopment stages is importantNew applications will rely on RT Ethernet ascommunication technologySuitable methodology is needed to validatedistributed applications

RT Ethernet Residual Bus Simulation enables earlytestingCombination of model-based testing principles tovalidate non-functional requirements

4 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REAgenda

1 Motivation & Introduction

2 Background

3 Real-time Ethernet Residual Bus Simulation

4 Application & Results

5 Conclusion & Outlook

5 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REModel-based Testing Approach

The automotive development process is model drivenModels are utilized as specifications for

Representing implementation detailsModeling system requirements

Test cases are systematically inherited from modelsExecution of cases on different test platforms

MiL, SiL, PiL, HiL and Residual Bus Simulation

6 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REResidual Bus Simulation

The remaining network is simulated from theviewpoint of the SUTSUT and simulator are coupled via thecommunication interfaceBehavior and network specific characteristics arerealistically emulatedThe simulator pretends to be a physical system

7 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

RERT Ethernet as In-Car NetworkAttributes of TTEthernet

TTEthernet provides three different message classes

BE0

Multimedia

BE BE

t

BE BE BE

BE Best-Effort Message

t

RCHeadlight

RC

0

RC Rate-Constrained Message

TTE-Switch

TTE-Switch

t

Chassis

TTTT

0 Cycle

Cycle0

t

TTTT

TT Time-Triggered Message

Static designed routing for deterministic behaviorSynchronized time base for time-triggeredcommunication

8 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REAgenda

1 Motivation & Introduction

2 Background

3 Real-time Ethernet Residual Bus Simulation

4 Application & Results

5 Conclusion & Outlook

9 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REModel-based MethodologyOverview

Requirements

Models

DownloadTest Case

Generation

T 1 s 2 s 5 s 7 s 9 s 11 sU u1 = HL_OFF u1 = LED_0 u1 = LED_100 u1 = LED_50 u1 = LED_101 u1 = LED_75

Y y1 = HL_OFF y1 = LED_0 y1 = LED_100 y1 = LED_50 y1 = LED_50 y1 = LED_75

Yact y1 = HL_OFF y1 = LED_0 y1 = LED_100 y1 = LED_50 y1 = LED_50 y1 = LED_75

L l1(u1, y1) = 500 µs�L jL1(l1) 100 µsLact l1(u1, y1) = 517 µs to 518 µs, MED = 518 µs, AVG = 518 µsR r1(y1) = 5000 µs�R jR1(r1) 10 µsRact r1(y1) = 4998 µs to 5002 µs, MED = 5000 µs, AVG = 5000µs

Table 3: Tested application with positive test result. Dark gray cells depict negative, light gray positive test results

7. REFERENCES[1] Aeronautical Radio Incorporated. Aircraft Data

Network, Part 7, Avionics Full-Duplex SwitchedEthernet Network. Standard ARINC Report 664P7-1,ARINC, 2009.

[2] E. Bringmann and A. Kramer. Model-Based Testingof Automotive Systems. In 2008 1st Int. Conf. onSoftware Testing, Verification and Validation, pages485–493, Apr. 2008.

[3] S. Demathieu, F. Thomas, C. Andre, S. Gerard, andF. Terrier. First Experiments Using the UML Profilefor MARTE. In 11th IEEE Int. Symp. on ObjectOriented Real-Time Distributed Computing 2008,pages 50–57, Piscataway, New Jersey, May 2008. IEEEPress.

[4] Eberspacher. FlexConfig RBS - Remaining BusSimulation for FlexRay and CAN with FlexConfigRBS. http://www.eberspaecher-electronics.com, 2014.

[5] FlexRay Consortium. FlexRay CommunicationsSystem Electrical Physical Layer Specification.Specification 3.0.1, FlexRay Consortium, Stuttgart,Oct. 2010.

[6] T. M. Galla. Cluster Simulation in Time-TriggeredReal-Time Systems. PhD-Thesis, TU Wien, Dec. 1999.

[7] International Organization for Standardization. Roadvehicles – Controller Area Network (CAN). ISO11898, ISO, Genf, 2003.

[8] O. Karfich, F. Bartols, T. Steinbach, F. Korf, andT. C. Schmidt. A Hardware/Software Platform forReal-time Ethernet Cluster Simulation in OMNeT++.In Proc. of the 6th Int. ICST Conf. on SimulationTools and Techniques, pages 334–337, New York, Mar.2013. ACM-DL.

[9] MOST Cooperation. MOST Specification Rev. 3.0 E2.Technical report, MOST, July 2010.

[10] K. Muller, T. Steinbach, F. Korf, and T. C. Schmidt.A Real-time Ethernet Prototype Platform forAutomotive Applications. In 2011 IEEE Int. Conf. onConsumer Electronics - Berlin, pages 221–225,Piscataway, New Jersey, Sept. 2011. IEEE Press.

[11] Object Management Group. MARTE - Modeling AndAnalysis Of Real-Time Embedded Systems.

[12] Object Management Group. UML - Unified ModelingLanguage.

[13] OMNeT++ Community. OMNeT++ 4.4.1.

[14] A. Pretschner, M. Broy, I. H. Kruger, and T. Stauner.Software Engineering for Automotive Systems: ARoadmap. In 2007 Future of Software Engineering,

FOSE ’07, pages 55–71, Washington, DC, USA, May2007. IEEE Computer Society.

[15] P. Skruch, M. Panek, and B. Kowalczyk. Model-BasedTesting in Embedded Automotive Systems. InJ. Zander, I. Schiefdecker, and P. J. Mosterman,Editors, Model-Based Testing for Embedded Systems,Computational Analysis, Synthesis and Design ofDynamic Systems Series, Chap. 19, pages 545–578.CRC Press, Boca Raton, Florida, Sept. 2011.

[16] Society of Automotive Engineers - AS-2D TimeTriggered Systems and Architecture Committee.Time-Triggered Ethernet AS6802. SAE Aerospace,Nov. 2011.

[17] T. Steinbach, H. Dieumo Kenfack, F. Korf, and T. C.Schmidt. An Extension of the OMNeT++ INETFramework for Simulating Real-time Ethernet withHigh Accuracy. In Proc. of the 4th Int. ICST Conf. onSimulation Tools and Techniques, pages 375–382, NewYork, Mar. 2011. ACM-DL.

[18] T. Steinbach, F. Korf, and T. C. Schmidt. Real-timeEthernet for Automotive Applications: A Solution forFuture In-Car Networks. In 2011 IEEE Int. Conf. onConsumer Electronics - Berlin, pages 216–220,Piscataway, New Jersey, Sept. 2011. IEEE Press.

[19] T. Steinbach, H.-T. Lim, F. Korf, T. C. Schmidt,D. Herrscher, and A. Wolisz. Tomorrow’s In-CarInterconnect? A Competitive Evaluation of IEEE802.1 AVB and Time-Triggered Ethernet (AS6802). In2012 IEEE Vehicular Technology Conf. (VTC Fall),Piscataway, New Jersey, Sept. 2012. IEEE Press.

[20] W. Steiner and G. Bauer. Mixed-criticality Networksfor Adaptive Systems. In 2010 IEEE/AIAA 29thDigital Avionics Systems Conf. (DASC), pages5.A.3–1–5.A.3–10, Oct. 2010.

[21] M. Utter, A. Pretschner, and B. Legeard. ATaxonomy of Modelbased Testing. Technical ReportWorking Paper: 04/2006, University of Waikato,Hamilton, New Zealand, Apr. 2006.

[22] Vector Informatik. CANoe 8.2 - ECU Development &Test with CANoe. http://vector.com, 2014.

[23] J. Zander, I. Schiefdecker, and P. J. Mosterman. ATaxonomy of Modelbased Testing for EmbeddedSystems from Multiple Industry Domains. InJ. Zander, I. Schiefdecker, and P. J. Mosterman,editors, Model-Based Testing for Embedded Systems,Computational Analysis, Synthesis and Design ofDynamic Systems Series, Chap. 1, pages 3–22. CRCPress, Boca Raton, Florida, Sept. 2011.

Test Cases

T 1 s 2 s 5 s 7 s 9 s 11 sU u1 = HL_OFF u1 = LED_0 u1 = LED_100 u1 = LED_50 u1 = LED_101 u1 = LED_75

Y y1 = HL_OFF y1 = LED_0 y1 = LED_100 y1 = LED_50 y1 = LED_50 y1 = LED_75

Yact y1 = HL_OFF y1 = LED_0 y1 = LED_100 y1 = LED_50 y1 = LED_50 y1 = LED_75

L l1(u1, y1) = 500 µs�L jL1(l1) 100 µsLact l1(u1, y1) = 517 µs to 518 µs, MED = 518 µs, AVG = 518 µsR r1(y1) = 5000 µs�R jR1(r1) 10 µsRact r1(y1) = 4998 µs to 5002 µs, MED = 5000 µs, AVG = 5000µs

Table 3: Tested application with positive test result. Dark gray cells depict negative, light gray positive test results

7. REFERENCES[1] Aeronautical Radio Incorporated. Aircraft Data

Network, Part 7, Avionics Full-Duplex SwitchedEthernet Network. Standard ARINC Report 664P7-1,ARINC, 2009.

[2] E. Bringmann and A. Kramer. Model-Based Testingof Automotive Systems. In 2008 1st Int. Conf. onSoftware Testing, Verification and Validation, pages485–493, Apr. 2008.

[3] S. Demathieu, F. Thomas, C. Andre, S. Gerard, andF. Terrier. First Experiments Using the UML Profilefor MARTE. In 11th IEEE Int. Symp. on ObjectOriented Real-Time Distributed Computing 2008,pages 50–57, Piscataway, New Jersey, May 2008. IEEEPress.

[4] Eberspacher. FlexConfig RBS - Remaining BusSimulation for FlexRay and CAN with FlexConfigRBS. http://www.eberspaecher-electronics.com, 2014.

[5] FlexRay Consortium. FlexRay CommunicationsSystem Electrical Physical Layer Specification.Specification 3.0.1, FlexRay Consortium, Stuttgart,Oct. 2010.

[6] T. M. Galla. Cluster Simulation in Time-TriggeredReal-Time Systems. PhD-Thesis, TU Wien, Dec. 1999.

[7] International Organization for Standardization. Roadvehicles – Controller Area Network (CAN). ISO11898, ISO, Genf, 2003.

[8] O. Karfich, F. Bartols, T. Steinbach, F. Korf, andT. C. Schmidt. A Hardware/Software Platform forReal-time Ethernet Cluster Simulation in OMNeT++.In Proc. of the 6th Int. ICST Conf. on SimulationTools and Techniques, pages 334–337, New York, Mar.2013. ACM-DL.

[9] MOST Cooperation. MOST Specification Rev. 3.0 E2.Technical report, MOST, July 2010.

[10] K. Muller, T. Steinbach, F. Korf, and T. C. Schmidt.A Real-time Ethernet Prototype Platform forAutomotive Applications. In 2011 IEEE Int. Conf. onConsumer Electronics - Berlin, pages 221–225,Piscataway, New Jersey, Sept. 2011. IEEE Press.

[11] Object Management Group. MARTE - Modeling AndAnalysis Of Real-Time Embedded Systems.

[12] Object Management Group. UML - Unified ModelingLanguage.

[13] OMNeT++ Community. OMNeT++ 4.4.1.

[14] A. Pretschner, M. Broy, I. H. Kruger, and T. Stauner.Software Engineering for Automotive Systems: ARoadmap. In 2007 Future of Software Engineering,

FOSE ’07, pages 55–71, Washington, DC, USA, May2007. IEEE Computer Society.

[15] P. Skruch, M. Panek, and B. Kowalczyk. Model-BasedTesting in Embedded Automotive Systems. InJ. Zander, I. Schiefdecker, and P. J. Mosterman,Editors, Model-Based Testing for Embedded Systems,Computational Analysis, Synthesis and Design ofDynamic Systems Series, Chap. 19, pages 545–578.CRC Press, Boca Raton, Florida, Sept. 2011.

[16] Society of Automotive Engineers - AS-2D TimeTriggered Systems and Architecture Committee.Time-Triggered Ethernet AS6802. SAE Aerospace,Nov. 2011.

[17] T. Steinbach, H. Dieumo Kenfack, F. Korf, and T. C.Schmidt. An Extension of the OMNeT++ INETFramework for Simulating Real-time Ethernet withHigh Accuracy. In Proc. of the 4th Int. ICST Conf. onSimulation Tools and Techniques, pages 375–382, NewYork, Mar. 2011. ACM-DL.

[18] T. Steinbach, F. Korf, and T. C. Schmidt. Real-timeEthernet for Automotive Applications: A Solution forFuture In-Car Networks. In 2011 IEEE Int. Conf. onConsumer Electronics - Berlin, pages 216–220,Piscataway, New Jersey, Sept. 2011. IEEE Press.

[19] T. Steinbach, H.-T. Lim, F. Korf, T. C. Schmidt,D. Herrscher, and A. Wolisz. Tomorrow’s In-CarInterconnect? A Competitive Evaluation of IEEE802.1 AVB and Time-Triggered Ethernet (AS6802). In2012 IEEE Vehicular Technology Conf. (VTC Fall),Piscataway, New Jersey, Sept. 2012. IEEE Press.

[20] W. Steiner and G. Bauer. Mixed-criticality Networksfor Adaptive Systems. In 2010 IEEE/AIAA 29thDigital Avionics Systems Conf. (DASC), pages5.A.3–1–5.A.3–10, Oct. 2010.

[21] M. Utter, A. Pretschner, and B. Legeard. ATaxonomy of Modelbased Testing. Technical ReportWorking Paper: 04/2006, University of Waikato,Hamilton, New Zealand, Apr. 2006.

[22] Vector Informatik. CANoe 8.2 - ECU Development &Test with CANoe. http://vector.com, 2014.

[23] J. Zander, I. Schiefdecker, and P. J. Mosterman. ATaxonomy of Modelbased Testing for EmbeddedSystems from Multiple Industry Domains. InJ. Zander, I. Schiefdecker, and P. J. Mosterman,editors, Model-Based Testing for Embedded Systems,Computational Analysis, Synthesis and Design ofDynamic Systems Series, Chap. 1, pages 3–22. CRCPress, Boca Raton, Florida, Sept. 2011.

T 1 s 2 s 5 s 7 s 9 s 11 sU u1 = HL_OFF u1 = LED_0 u1 = LED_100 u1 = LED_50 u1 = LED_101 u1 = LED_75

Y y1 = HL_OFF y1 = LED_0 y1 = LED_100 y1 = LED_50 y1 = LED_50 y1 = LED_75

Yact y1 = HL_OFF y1 = LED_0 y1 = LED_100 y1 = LED_50 y1 = LED_50 y1 = LED_75

L l1(u1, y1) = 500 µs�L jL1(l1) 100 µsLact l1(u1, y1) = 517 µs to 518 µs, MED = 518 µs, AVG = 518 µsR r1(y1) = 5000 µs�R jR1(r1) 10 µsRact r1(y1) = 4998 µs to 5002 µs, MED = 5000 µs, AVG = 5000µs

Table 3: Tested application with positive test result. Dark gray cells depict negative, light gray positive test results

7. REFERENCES[1] Aeronautical Radio Incorporated. Aircraft Data

Network, Part 7, Avionics Full-Duplex SwitchedEthernet Network. Standard ARINC Report 664P7-1,ARINC, 2009.

[2] E. Bringmann and A. Kramer. Model-Based Testingof Automotive Systems. In 2008 1st Int. Conf. onSoftware Testing, Verification and Validation, pages485–493, Apr. 2008.

[3] S. Demathieu, F. Thomas, C. Andre, S. Gerard, andF. Terrier. First Experiments Using the UML Profilefor MARTE. In 11th IEEE Int. Symp. on ObjectOriented Real-Time Distributed Computing 2008,pages 50–57, Piscataway, New Jersey, May 2008. IEEEPress.

[4] Eberspacher. FlexConfig RBS - Remaining BusSimulation for FlexRay and CAN with FlexConfigRBS. http://www.eberspaecher-electronics.com, 2014.

[5] FlexRay Consortium. FlexRay CommunicationsSystem Electrical Physical Layer Specification.Specification 3.0.1, FlexRay Consortium, Stuttgart,Oct. 2010.

[6] T. M. Galla. Cluster Simulation in Time-TriggeredReal-Time Systems. PhD-Thesis, TU Wien, Dec. 1999.

[7] International Organization for Standardization. Roadvehicles – Controller Area Network (CAN). ISO11898, ISO, Genf, 2003.

[8] O. Karfich, F. Bartols, T. Steinbach, F. Korf, andT. C. Schmidt. A Hardware/Software Platform forReal-time Ethernet Cluster Simulation in OMNeT++.In Proc. of the 6th Int. ICST Conf. on SimulationTools and Techniques, pages 334–337, New York, Mar.2013. ACM-DL.

[9] MOST Cooperation. MOST Specification Rev. 3.0 E2.Technical report, MOST, July 2010.

[10] K. Muller, T. Steinbach, F. Korf, and T. C. Schmidt.A Real-time Ethernet Prototype Platform forAutomotive Applications. In 2011 IEEE Int. Conf. onConsumer Electronics - Berlin, pages 221–225,Piscataway, New Jersey, Sept. 2011. IEEE Press.

[11] Object Management Group. MARTE - Modeling AndAnalysis Of Real-Time Embedded Systems.

[12] Object Management Group. UML - Unified ModelingLanguage.

[13] OMNeT++ Community. OMNeT++ 4.4.1.

[14] A. Pretschner, M. Broy, I. H. Kruger, and T. Stauner.Software Engineering for Automotive Systems: ARoadmap. In 2007 Future of Software Engineering,

FOSE ’07, pages 55–71, Washington, DC, USA, May2007. IEEE Computer Society.

[15] P. Skruch, M. Panek, and B. Kowalczyk. Model-BasedTesting in Embedded Automotive Systems. InJ. Zander, I. Schiefdecker, and P. J. Mosterman,Editors, Model-Based Testing for Embedded Systems,Computational Analysis, Synthesis and Design ofDynamic Systems Series, Chap. 19, pages 545–578.CRC Press, Boca Raton, Florida, Sept. 2011.

[16] Society of Automotive Engineers - AS-2D TimeTriggered Systems and Architecture Committee.Time-Triggered Ethernet AS6802. SAE Aerospace,Nov. 2011.

[17] T. Steinbach, H. Dieumo Kenfack, F. Korf, and T. C.Schmidt. An Extension of the OMNeT++ INETFramework for Simulating Real-time Ethernet withHigh Accuracy. In Proc. of the 4th Int. ICST Conf. onSimulation Tools and Techniques, pages 375–382, NewYork, Mar. 2011. ACM-DL.

[18] T. Steinbach, F. Korf, and T. C. Schmidt. Real-timeEthernet for Automotive Applications: A Solution forFuture In-Car Networks. In 2011 IEEE Int. Conf. onConsumer Electronics - Berlin, pages 216–220,Piscataway, New Jersey, Sept. 2011. IEEE Press.

[19] T. Steinbach, H.-T. Lim, F. Korf, T. C. Schmidt,D. Herrscher, and A. Wolisz. Tomorrow’s In-CarInterconnect? A Competitive Evaluation of IEEE802.1 AVB and Time-Triggered Ethernet (AS6802). In2012 IEEE Vehicular Technology Conf. (VTC Fall),Piscataway, New Jersey, Sept. 2012. IEEE Press.

[20] W. Steiner and G. Bauer. Mixed-criticality Networksfor Adaptive Systems. In 2010 IEEE/AIAA 29thDigital Avionics Systems Conf. (DASC), pages5.A.3–1–5.A.3–10, Oct. 2010.

[21] M. Utter, A. Pretschner, and B. Legeard. ATaxonomy of Modelbased Testing. Technical ReportWorking Paper: 04/2006, University of Waikato,Hamilton, New Zealand, Apr. 2006.

[22] Vector Informatik. CANoe 8.2 - ECU Development &Test with CANoe. http://vector.com, 2014.

[23] J. Zander, I. Schiefdecker, and P. J. Mosterman. ATaxonomy of Modelbased Testing for EmbeddedSystems from Multiple Industry Domains. InJ. Zander, I. Schiefdecker, and P. J. Mosterman,editors, Model-Based Testing for Embedded Systems,Computational Analysis, Synthesis and Design ofDynamic Systems Series, Chap. 1, pages 3–22. CRCPress, Boca Raton, Florida, Sept. 2011.

Stimuli

Reaction SUTTest Case Execution

Test Data

SUT Data

Requirements are modeled within suitable diagramsTest cases are inherited from the diagramsTest cases are executed on a suitable residual bussimulation platform

10 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REModeling System RequirementsUML-MARTE

Classic UML is not sufficient for embeddedReal-time SystemsUtilization of UML-Profile Modeling and Analysis ofReal-time Embedded Systems (MARTE)

sd MessageProcessing <<TimedConstrained>>on = [MARTE_Library::TimedLibrary::idealClock]

<<timedConstrained>>{(messageReceived - replySent) = (500,µs)}{(jitter(messageReceived - replySent)) ≤ (100,µs)}

Dashboard-System Headlight-Controller

setNewLightStatus(status)

@messageReceived

@replySent

sendCurrentLightStatus(status)

11 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REAbstract Test CasesDefinition

Base Model

System y(t)u(t)

ATCFR = (T ,U,Y )

Modeling specific values ofinputs and outputs atspecific points in time

Extending the Model

System y(t)u(t)

Latency

Rate

t1 t2t3

ATCNFR =(T ,U,Y , L,R,∆L,∆R)

Extending with reply time(latency) & transmission rate(rate)

12 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REAbstract Test CasesUtilization

Abstract representation of to be generated test dataModeling functional requirements with expectedoutputModeling non-functional requirements with expectedtiming constraintsUtilization as simulation model to drive the simulator

13 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REImplementation of our ApproachRequirements and Architecture

RequirementsTTEthernet compliant message transmissionSupport of timing analyzesExecution of the abstract test case model

ConfiguratorUML-MARTE

Models

XML-Doc

Abstract Test Case

Residual Bus Simulator

DPM

SUT

EthernetFrames

14 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REAgenda

1 Motivation & Introduction

2 Background

3 Real-time Ethernet Residual Bus Simulation

4 Application & Results

5 Conclusion & Outlook

15 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REOverview of the physical system

Drive-by-wire (W) Switch Switch Drive-by-

wire (SW)

Camera Multimedia Display

Headlight L Headlight R

Lightcontrol Dashboard

MultimediaDashboard

SuT

to be simulated by the residual bus simulator

Light control dashboard transmits new light statesHeadlights reply each received light state andPeriodically provide their current light stateLight control dashboard presents the light state tothe user

16 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REValidating the Headlight ControllerRequirement Modelling with UML-MARTE

sd MessageProcessing <<TimedConstrained>>on = [MARTE_Library::TimedLibrary::idealClock]

<<timedConstrained>>{(messageReceived - replySent) = (500,µs)}{(jitter(messageReceived - replySent)) ≤ (100,µs)}

Dashboard-System Headlight-Controller

setNewLightStatus(status)

@messageReceived

@replySent

sendCurrentLightStatus(status)

Timing requirements of the reply messageLatency: 500 µsJitter: ± 50 µs

17 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REValidating the Headlight ControllerRequirement Modelling with UML-MARTE

sd TimedStateTransmission

<<TimedConstrained>>on = [MARTE_Library::TimedLibrary::idealClock]

Dashboard-SystemHeadlight-Controller

@ sentStatusi

updateCurrentLightstatus(status)

<<timedConstrained>>{(sendStatusi - sendStatusi-1) = (5000,µs)}{(jitter(sendStatusi - sendStatusi-1)) ≤ (10,µs)}

Repeat[“timedEvent” (5,ms)]

<<TimedProcessing>>on = [MARTE_Library::TimedLibrary::idealClock]

Timing requirements of the message transmissionRate: 5000 µsJitter: ± 5 µs

18 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REValidating the Headlight Controller

T 1 s 2 s 5 s 7 s 9 s 11 sU u1 = HL_OFF u1 = LED_0 u1 = LED_100 u1 = LED_50 u1 = LED_101 u1 = LED_75Y y1 = HL_OFF y1 = LED_0 y1 = LED_100 y1 = LED_50 y1 = LED_50 y1 = LED_75Yact y1 = HL_OFF y1 = HL_OFF y1 = HL_OFF y1 = HL_OFF y1 = HL_OFF y1 = HL_OFFL l1(u1, y1) = 500 µs

∆L jL1(l1) ≤ 100 µsLact l1(u1, y1) = 518 µs to 518 µs, MED = 518 µs, AVG = 518 µsR r1(y1) = 5000 µs

∆R jR1(r1) ≤ 10 µsRact r1(y1) = 4998 µs to 5002 µs, MED = 5000 µs, AVG = 5000 µs

Functional requirements cannot be fulfilledExpected values are not located at the outputNon-functional timing requirement are fulfilledLatency of the acknowledgement lay within theallowed range

19 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REAgenda

1 Motivation & Introduction

2 Background

3 Real-time Ethernet Residual Bus Simulation

4 Application & Results

5 Conclusion & Outlook

20 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REConclusion

Residual bus simulator is directly connected with theSUTMessage classes and a synchronization procedure aresupportedNon-functional timing requirements are modeledwithin UML-MARTEAbstract test case model models functional andnon-functional test dataUtilization of abstract test cases as simulation modelSuccessful utilization for the validation of an RTEthernet application

21 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

REOutlook

Investigate how AUTOSAR and EAST-ADL couldco-exist with our approachImplement a RBS with a more suitable architecturewithout dual-port memoryAnalyze the real-time and performance aspects ofthe new architecture

22 / 23

Hochschule für Angewandte Wissenschaften Hamburg Hamburg University of Applied Sciences

RT-Ethernet ResidualBus Simulation

F. Bartols

Motivation &Introduction

Background

RT Ethernet RBS

Application & Results

Conclusion & Outlook

RE

Thank you for your attention

Visit us at:https://core.informatik.haw-hamburg.de

23 / 23