Model-Based Design for Automotive Systems

Model-Based Design for Automotive C t l S tControl Systems h W o r k s , I n c . © 2 0 1 0 M a t h Agenda  Overview of The MathWorks  Introduction of Model-Based Design  Key MathWorks Products y  Early Verification and Validation Tools  RP and HIL Solution 2 The MathWorks at a Glance Headquarters: Natick, Massachusetts USA USAUSA: California, Michigan, Washington DC, Texas Europe: UK, France, Germany, Switzerland, Italy, Spain, the Netherlands, Sweden Asia-Pacific: Chi K A t li J I diChina, Korea, Australia, Japan, India Worldwide training and consulting Di t ib t i 25 t i Earth’s topography on an equidistant cylindrical projection, Distributors in 25 countries created with MATLAB and Mapping Toolbox 3 MathWorks Today  Revenues ~$500M in 2008  Privately held Privately held  More than 2,200 employees worldwide  Worldwide revenue balance: 45% North America, 55% international  More than 1,000,000 users in 175+ countries 1985 1990 1995 2000 2005 4 Key Industries  Aerospace and Defense  Automotive  Biotech and Pharmaceutical  Chemical/Petrochemical  Communications  Education  Electronics  Financial Services  Industrial Automation and Machinery  Power and Energygy  Semiconductor 5 MathWorks Product Family Overview View full product list Simulink Product Family Application Specific ProductsSimulink Product Family Application-Specific Products 6 MATLAB Product Family Agenda  Overview of The MathWorks  Introduction of Model-Based Design  Key MathWorks Products y  Early Verification and Validation Tools  RP and HIL Solution 7 Current Challenges in Automotive  Environmental issuesEnvironmental issues  Oil shortage and resulting price development  Legal regulations  CO2 Emission Limits; e.g. 120g/km by 2015 in EU; g g y Customer Requirements Customer Requirements  Cost efficient and save vehicles 88 Current Challenges in Automotive Applications under development Lights Climate Controls Voice Recognition Applications under development Power Engine Control Obstacle Windows Lights Driver Drowsiness Infotainment Instrumentation Recognition Navigation Power Management Transmission Control Steering Obstacle Detection Crash Avoidance Airbags Passenger Doors Instrumentation  2010: Electronics expected to comprise 40% of material cost (SAE AEI March 2005) Co t o Ride Control ABS Stability Controls Avoidance Adaptive Front Detection Wireless Connectivity Traction Control Adaptive Cruise Control Lighting Systems Tire Pressure Monitoring 9 Current Challenges in Automotive Applications under developmentApplications under development Hybrid Electrical Vehicles Brake-Energy Pure Electrical Vehicles Brake Energy Recuperation Start-Stop Systems Battery Management SystemsSystems Fuel Cells Electrical Motors Fuel Cells Electrical Power Electrical 1010 Steering Accessories Explosive Growth of Code Estimates for Lines of Code (LOC)  Today’s powertrain: 500,000 LOC “Growth of top end automotive embedded software has beeny p ,  Today’s vehicles: 1,000,000 LOC  2015 vehicles: 100,000,000 LOC embedded software has been exponential.” Robert Gee Director of Strategy for Motorola Automotive * 2006 figures from SAE AEI Director of Strategy for Motorola Automotive “…no longer possible to lid t d ifvalidate and verify functionality by brute force testing alone.” Jim Kolhoff Automotive Engineering “Managing for Jim Kolhoff Director of Software Engineering at GM Powertrain 11 Automotive Engineering, Managing for Software Success” – Aug 2006 sae.org/automag/electronics/08-2006/1-114-8-34.pdf Model Based DesignModel Based Design Requirements Vehicle Integration & Calibration Simulation Hardware-in-the-Loop Testing System Design Hardware/Software Integration Verification and Validation HIL xPC Processor-in-the-Loop Testing Syste es g Integration Rapid PrototypingMatlab/Simulink/Stateflow Design Verifier xPC PIL On-Target Rapid Prototyping Software-in-the-Loop Testing Software Design SoftwareIntegration Design Verifier Physical Modeling PIL Target Support Coding PolySpace SIL Fixed-Point RTW/E-Coder 12 Production Code Generation g SIL Target Support P bl ith T diti l D l tProblems with Traditional Development Design ImplementationRequirements Test and Design Implementationand Specs Verification Text Physical Manual Traditional testingText documents prevents rapid iteration Physical prototypes incomplete, expensive Manual implementation separate tools & human error Traditional testing errors found late in process 13 A l i M d l B d D iApplying Model-Based Design Design ImplementationRequirements Test and Design Implementationand Specs Verification Physical Manual Traditional testingPhysical prototypes incomplete, expensive Manual implementation separate tools & human error Traditional testing errors found late in process Executable Specification • Unambiguous spec, supplemented by text • One set of models for all teams 14 O e set o ode s o a tea s • Model whole system including environment • Block diagram description • Early validation and test development A l i M d l B d D iApplying Model-Based Design Design ImplementationRequirements Test and Design Implementationand Specs Verification Manual Traditional testingManual implementation separate tools & human error Traditional testing errors found late in process Design with Simulation • Systematic design exploration and optimization Fi d fl b f i l t ti 15 • Find flaws before implementation • Apply to both controller and physical plant • Incremental design from system level to implementation A l i M d l B d D iApplying Model-Based Design Design ImplementationRequirements Test and Design Implementationand Specs Verification Traditional testingTraditional testing errors found late in process Automatic Code GenerationAutomatic Code Generation • No manual coding errors • Hardware target portability • Improved testability due to repeatability Bridge between domain software and hardware 16 • Bridge between domain, software and hardware knowledge • Hardware-in-Loop for physical model A l i M d l B d D iApplying Model-Based Design Design ImplementationRequirements Test and Design Implementationand Specs Verification Model elaboration Continuous verification Continuous Test and Verification • Detect errors early in development R d d d h i l t t 17 • Reduce dependency on physical prototypes • Implementations that work the first time • Reuse test suites across development stages Requirements Test and Verification Model-Based Design Workflow Requirements and Specs Verification Implementation Design 18 The Val e of Model Based DesignThe Value of Model-Based Design Model-Based Design Innovation  Executable specification  Design with simulation  Implementation through code ti  Rapid design iterations  “What-if” studies  Unique features and differentiators generation  Continuous test and verification Quality Reduce design errors  Minimize hand coding errors U bi i ti i t ll Unambiguous communication internally and externally Cost  Reduce expensive physical prototypes  Reduce re-work  Automate testing Time-to-market  Get it right the first time 19 Agenda  Overview of The MathWorks  Introduction of Model-Based Design  Key MathWorks Products y  Early Verification and Validation Tools  RP and HIL Solution 20 MathWorks provide MBD tools and Physical modeling tools to integrate control software design and control target modeling to in a common platforma common platform s h o p f i e r t s t l i d a t i o n e m s l i n e a n i c s u l i c s o n i c s m e W o r k s i g n V e r i f x e d P o i n t s t e m t e s s n a n d V a l o w e r S y s t S i m D r i v e S i m M e c h a S i m H y d r a u i m E l e c t r o e R e a l T i D e s i F i x S y s V e r i f i c a t i o S i m P o Simscape S S S S i P o l y s p a c e MBC Simulink V Simscape P MBC 21 MATLABMATLAB Core MathWorks Products The leading environment forg technical computing – The de facto industry-standard, high-level programming language for algorithm development – Numeric computation – Data analysis and visualization – Toolboxes for signal and image processing, statistics, optimization, b li th d thsymbolic math, and other areas – Foundation of MathWorks products 22 Simulink  System-level modeling – MultidomainMultidomain – Graphical – Interactive – Hierarchical  Algorithm design  Simulation Simulation – Model is an “executable specification” 23 St t flStateflow  Extend Simulink with a design environment for developing Extend Simulink with a design environment for developing state machines and flow charts  Design systems containing control, supervisory, and mode g y g , p y, logic  Describe logic in a natural and understandable form with d t i i ti ti tideterministic execution semantics 24 Implement Design with Code Generation Real-Time Workshop Embedded CoderReal Time Workshop Embedded Coder  Generate ANSI/ISO C code from Simulink models – Readable – Traceable C t– Compact – Efficient – Consistent – For deployment to microprocessors and DSPs /* Outport: '/heat' incorporates: * Product: '/Divide'  Generating HDL Code – For deployment to FPGAs and ASICs * Product: /Divide * Inport: '/Sensor' */ rtY.heat = (int16_T)(rtU.Sensor >> 5); 25 FPGAs and ASICs Agenda  Overview of The MathWorks  Introduction of Model-Based Design  Key MathWorks Products y  Early Verification and Validation Tools  RP and HIL Solution 26 Tracing RequirementsModel Simulink Verification and Validation Creating links between textual documents and model objectsj 27 Tracing RequirementsSource Code Simulink Verification and Validation R l Ti W k h E b dd d C dReal-Time Workshop Embedded Coder Including requirements in the generated source code 28 Tracing ModelSource Code Real-Time Workshop Embedded Coder Bidirectional navigation between the model and the generated codegenerated code 29 Traceability – Summary Simulink Verification and Validation Real-Time Workshop Embedded CoderReal Time Workshop Embedded Coder Requirements ModelRequirements Model Code Requirements Traceability – Report Traceability Report Code 30 Simulink Verification and Validation Real-Time Workshop Embedded Coder ISO 26262 工具资质认证 首个具有ISO/DIS 26262资质的产品代 码生成器成 Real-Time Workshop Embedded Coder和PolySpace已经通过TÜV SÜD 的资质验证的资质验证 MathWorks创建的工具认证包和开发 流程 快递问题件怎么处理流程河南自建厂房流程下载关于规范招聘需求审批流程制作流程表下载邮件下载流程设计 也通过了TÜV SÜD的评估 用户通过对工具认证包的客户化及参 考相应的证 书 关于书的成语关于读书的排比句社区图书漂流公约怎么写关于读书的小报汉书pdf /认证 报告 软件系统测试报告下载sgs报告如何下载关于路面塌陷情况报告535n,sgs报告怎么下载竣工报告下载 来进行工具资 质认证 包含如下模板: 软件工具资质 计划 项目进度计划表范例计划下载计划下载计划下载课程教学计划下载 软件 具文档 3131 软件工具文档 软件工具分类 分析 定性数据统计分析pdf销售业绩分析模板建筑结构震害分析销售进度分析表京东商城竞争战略分析 软件工具资质报告 ISO 26262 模型自动校验 32 MISRA-C:2004 自动校验 33 模型测试 Simulink Design VerifierSimulink Design Verifier  Building exhaustive tests is hard and time consuming – Example: Getting 100% model coverage  Exhaustive structural testing is particularly important for Exhaustive structural testing is particularly important for – Safety critical applications – Complex logic C t b d d l t– Component based development  Generated test cases are reusable – Test design in the model form – Test code after implementation Test code on target processor 34 – Test code on target processor 模型覆盖率测试 Simulink Verification and ValidationSimulink Verification and Validation  Model Coverage tool reports coverage metrics 35 Agenda  Overview of The MathWorks  Introduction of Model-Based Design  Key MathWorks Products y  Early Verification and Validation Tools  RP and HIL Solution 36  Models designed in Simulink and Stateflow From Simulation to Real-Time Testing Models designed in Simulink and Stateflow  You want to run, test, and prove your Simulink design with your hardware under test at its normal operating frequency, speed, or timing.g  But how can you do this testing in real-time? ? 37  Complete fully assembled real-time testing solution xPC Target  Complete, fully assembled, real time testing solution  Combines xPC Target (software) with a real-time target machine and IO modules (hardware)  RP, HIL 38 T d ’ C D d El i C lToday’s Car Depends on Electronic Controls Engine Management System CNG/LPG EMS Battery Management System Vehicle Controller Hybrid Delivery Truck Construction Equipment HVAC Common Rail Diesel Active Suspension Control Hybrid Powertrain Central Body Controller Telematics 3939 MathWorks Change the world by Accelerating the paceAccelerating the pace of discovery, innovation, development, and learning in engineering and science 40