數位轉型實踐之路:如何有系統地利用資料數據與模型來進行企業數位轉型

Pragmatic Digital Transformation Through the Systematic Use of Data and Models

摘要

目前許多進行數位轉型的企業,正從雄心勃勃的企圖心邁向付諸實現的階段。這些企業對於數位化轉型早已訂下了由上而下的高層級目標,正尋求工程師和科學家們的協助,希望藉由學習新的技術以及與全新但陌生的小組團隊合作,以提出新產品和服務來實現這些目標。

為了因應這一挑戰,技術型的企業必須能掌握如何去有系統地使用數據和模型,不僅在研發階段、所有跨團隊的整個產品生命週期也都要能跟得上才行,而且必須要考慮企業人員的本身技能、流程和科技的變化,才能稱得上是一個有效的數位轉型計畫。

本演講將描述何謂實用的數位化轉型,我們並將示範全球許多成功的工程和科學團隊們,如何利用數據和模型來進行數位轉型達成目標。

Organizations with digital transformation initiatives are making the transition from visionary ambitions to practical projects. These organizations have defined their high-level digital transformation objectives, and are now looking to their engineers and scientists to achieve them by learning new technologies, collaborating with unfamiliar groups, and proposing new products and services.

To meet this challenge, technical organizations must master how to systematically use data and models, not only during the research and development stages, but also across groups throughout the lifecycle of the offering. An effective digital transformation plan needs to consider changes in people's skills, processes, and technology.

Join us, this talk describes this pragmatic approach to digital transformation and demonstrates how engineering and scientific teams are leveraging data and models to achieve their transformative objectives.

Jim Tung

MathWorks Inc.

MATLAB & Simulink 2020a/2020b最新功能

What's New in 2020a/2020b of MATLAB & Simulink

摘要

本演講將介紹MATLAB & Simulink 於2020 a/b兩版本的最新強大功能如何支援研究、設計以及開發流程。2020a聚焦於深度學習、無線通訊、自動駕駛以及其他應用領域,您將可以了解新工具對於資料的預處理、分析;馬達控制演算法的開發、如何建立互動式的apps;如何打包模擬結果並將其分享出去,以及建模、模擬及驗證的設計流程等最新功能。

Learn about new capabilities in the MATLAB® and Simulink® product families to support your research, design, and development workflows. This talk highlights features for deep learning, wireless communications, automated driving, and other application areas. You will see new tools for preprocessing and analyzing data; developing motor control algorithms; creating interactive apps; packaging and sharing simulations; and modeling, simulating, and verifying designs.

Phoebe Li

TeraSoft Inc.

Jeffrey Liu

TeraSoft Inc.

機器學習:經過實證的應用與新功能

Machine Learning: Proven Applications and New Features

摘要

機器學習的應用無處不在。從自動化設計和發電,到醫療設備與生產製造,機器學習技術在日常生活的每個時刻,已經被用來進行重大的商業決策。

在本段演講,我們將聚焦於一些利用機器學習獲得卓越成果的應用,包含車隊資料的解析、能源預測、製造流程解析等等。我們也將展示工程師們如何把機器學習技術整合在他們的控制、訊號處理等工作流程中,進而提升系統效能。在整段演講之中,我們將特別介紹MATLAB的新功能,可幫助使用者更容易地開始使用機器學習、為本身的應用找到最佳模型、將機器學習模型與生產工作流程整合。其中包含了machine learning apps的更新、將自動化技巧運用在機器學習工作流程(AutoML)、產生機器學習模型的C/C++程式碼。透過本演講,您將了解其他人如何運用機器學習,且如何將機器學習應用在工作上並獲得啟發。

Agenda
    經過實證的應用:
  • 車隊資料解析
  • 能源預測
  • 生產資料解析
  • 新功能:
  • 可用來進行機器學習的MATLAB apps
  • AutoML (Automated Machine Learning,自動化機器學習)
  • 利用機器學習進行訊號處理
  • 將機器學習模型轉為C/C++程式碼

Machine learning is ubiquitous. From automotive design and electricity generation to medical devices and manufacturing, machine learning techniques are being used to make critical business decisions every moment of the day.

In this talk we will focus on applications where machine learning has enjoyed great success, including fleet data analysis, energy forecasting, and manufacturing process analytics. We will also demonstrate how engineers are integrating machine learning techniques with their controls and signal processing workflows to improve system performance.

Throughout the presentation we will highlight new features in MATLAB that make it easier to get started with machine learning, find the best model for your application, and integrate machine learning models with your production workflows. This includes updates to machine learning apps, applying automation techniques to the machine learning workflow (AutoML), and C/C++ code generation for machine learning models.

Come to this talk to learn how your peers have applied machine learning, and to get inspiration for how machine learning could be applied to your own work.

    Proven Applications:
  • Fleet Data Analytics
  • Energy Forecasting
  • Manufacturing Analytics
  • New Capabilities
  • MATLAB apps for Machine Learning
  • AutoML (Automated Machine Learning)
  • Signal Processing with Machine Learning
  • C/C++ Code Generation for Machine Learning Models

Abhijit Bhattacharjee

MathWorks Inc.

AI驅動系統中的關鍵資料 - 精準找到聲音中的關鍵字,以及後續處理作法

The Key Role of Data in Modern AI-Powered Systems - Spotting Voice Keywords and Beyond

摘要

當前有許多現代化的系統採用了機器學習與深度學習技術且不斷持續增加,並橫跨各項應用領域,包含最廣為人知的電腦視覺、訊號處理以及文字分析等。一般研究成果發表大致都可以有效地選出合適的深度學習模型 – 然而,這和開發實際運作的系統卻有很關鍵性的差異。現實中,大量用於模型訓練和評估的資料通常需要仰賴特定領域的資源、工具、和專家來整理及處理。

在本段演講,我們將透過實際範例藉由審視整個開發流程來探討這些必要條件 – 接著再將這些考量因素更廣泛地套用到不同應用中。範例中我們將利用MATLAB程式碼來探索,討論要做些什麼事,讓裝置可以被”Hey Siri”或”OK Google”等的觸發機制選擇性地喚醒;我們還特別提供一些針對特定資料的最佳實踐方法,包含資料標記和註記、資料攝入(data ingestion)、資料合成(synthesis)與增強(augmentation)、特徵擷取、以及轉域(domain transformation)等。

The adoption of machine learning and deep learning continues to grow in modern systems across many application areas, including most notably computer vision, signal processing, and text analytics. Research publications are generally effective for selecting suitable deep learning models - When developing real-work systems however, a key difference is often made by the application-specific resources, tools, and expertise for curating and processing the vast amounts of data used to train and evaluate those models.

In this session, we discuss those requirements by reviewing the development workflow of a practical example - We then generalize those considerations to a wider range of applications. Using MATLAB code, we explore what it takes to make a device selectively wake up with trigger phrases like "Hey Siri" or "OK Google". In particular, we cover a number of data-specific best practices, including for data labeling and annotation, data ingestion, data synthesis and augmentation, feature extraction, and domain transformations.

Abhijit Bhattacharjee

MathWorks Inc.

自動光學檢測與瑕疵偵測的產業應用

Automated Optical Inspection and Defect Detection for Industrial Applications

摘要

自動檢測與瑕疵偵測對於高產量生產系統的品質控管格外重要,這些技術被許多產業廣泛採納,用來檢測像是金屬鐵路、半導體晶圓、隱形眼鏡等製造的表面瑕疵,而近期深度學習的快速發展,使我們偵測瑕疵的能力得以大幅提升。在本段演講,你將了解如何利用MATLAB來開發,以深度學習為基礎可偵測不同種類異常現象、且在地化的方法。

演講焦點
  • 資料的取得與前處理技巧,包含除噪、定位(registration)、和灰階調整(intensity adjustment)
  • 瑕疵與異常的語意分析(semantic segmentation)和標記
  • 部署至多種硬體平台,如CPUs和GPUs

Automated inspection and defect detection are critical for high throughput quality control in production systems. They are widely adopted in many industries for detection of flaws on manufactured surfaces such as metallic rails, semiconductor wafers, contact lenses and so on. Recent developments in deep learning have significantly improved our ability to detect defects. In this session, you will learn how to use MATLAB to develop deep learning based approaches to detect and localize different types of anomalies. You will learn about:

  • Data access and preprocessing techniques including denoising, registration and intensity adjustment
  • Semantic segmentation and labeling of defects and abnormalities
  • Defect detection using MobileNetv2, Grad-CAM and other deep learning techniques
  • Deploying to multiple hardware platforms such as CPUs and GPUs

Abhijit Bhattacharjee

MathWorks Inc.

AI on Edge:以GPU Coder實現深度學習推理

AI on Edge:GPU Coder for Deep Learning Inference

摘要

GPU Coder 可將深度學習模型演算法從MATLAB Code轉換成CUDA C/C++原始碼,除了轉換模型外,也能將前處理的影像處理演算法與影像標記框,USB Camera api一起轉成CUDA Code,並且支援Yolov2與Yolov3,也可以與OpenCV算法整併,實現CUDA與C/C++/OpenCV的整合,並將MATLAB深度學習演算法建置在Jetson系列的嵌入式GPU中,如Nano,AGX,NX,TX2等,即可完成快速部屬實現於邊緣計算。

Fred Liu

TeraSoft Inc.

深度學習的嵌入式設備部署-當FPGAs準備好

Deploying Deep Learning on Embedded Devices – When FPGAs Make Sense

摘要

由於嵌入式設備固有的資源限制,設計深度學習、電腦視覺和訊號處理的應用,並將其部署到Xilinx Zynq™或NVIDIA®Jetson或ARM®處理器之類的FPGA,GPU和CPU平台上具有相當的挑戰性。本演講將引導您完成以MATLABR為基礎的深度學習部署之工作流程,並自動生成C / C ++或CUDAR或VHDL程式碼。

對於希望將深度學習整合到以FPGA為基礎應用的系統設計人員來說,本演講介紹了部署到FPGA硬體的一些挑戰和注意事項,我們將為您詳細介紹新發布的產品-【深度學習HDL工具箱】(Deep Learning HDL Toolbox™)的新功能,我們將展示如何使用來自MATLAB的產品預建的Bitstream,在FPGA上探索和原型化已被訓練好的深度學習網絡,您可以透過該產品進一步客製化自己的深度學習網絡,以符合性能要求和硬體資源使用情況、生成HDL,並將其整合到以FPGA為基礎的邊緣推斷(edge inference)系統中。

Designing deep learning, computer vision, and signal processing applications and deploying them to FPGAs, GPUs, and CPU platforms like Xilinx Zynq™ or NVIDIA® Jetson or ARM® processors is challenging because of resource constraints inherent in embedded devices. This talk walks you through a MATLABR based deployment workflow that generates C/C++ or CUDA® or VHDL code.

For system designers looking to integrate deep learning into their FPGA-based applications, the talk helps teach the challenges and considerations for deploying to FPGA hardware, and details the approach introduced with the newly released Deep Learning HDL Toolbox™. We will briefly show how to explore and prototype trained networks on FPGAs using prebuilt bistreams from MATLAB. You can further customize your network to meet your performance requirments and hardware resource usage, generate HDL, and integrate it into an FPGA-based edge inference system.

Phoebe Li

TeraSoft Inc.

各種MATLAB部署的方案: 以預測性維護應用為例

Various MATLAB Deployment Solutions for Predictive Maintenance

摘要

在本次演講中,我們將以預測性維護應用為例,展示各種MATLAB部署的方案,包含:

  • 獨立應用程式
  • Web App
  • 嵌入式系統
  • 與企業系統整合
  • 雲端

最後,我們將分享如何使用MATLAB相關工具從開發到部署的成功案例。

Judy Yang

TeraSoft Inc.

與MATLAB深度結合之軟體無線電平台

Highly-connected-with-MATLAB SDR Platform

摘要

本次演講主要介紹由工研院開發之軟體無線電平台,此平台主要特點就是使用上非常容易,使用者無需花太多時間了解硬體相關規格即可輕易的發送MATLAB所產生的I/Q raw data以及很輕易的接收到RF的I/Q raw data 進行分析。使用者只需專注在通訊理論上的研究,透過MATLAB強大的函式庫,可以實現over-the-air的實驗,寫得好的程式碼甚至也可以實現通訊儀器上的分析。

陳仁智

ITRI

以模型為基礎的FPGA, ASIC及SoC設計

Adopting Model-Based Design for FPGA, ASIC, and SoC

摘要

Model based design(MBD)基於模組設計利用Simulink進行數位硬體設計和驗證,擺脫親自編寫硬體描述語言,已經幫助了眾多客戶縮短其設計時程、提高了驗證的效率,並為他們的FPGA,ASIC或SoC設計流程中提供了更高品質的結果。儘管此一工作流程似乎與您現有的工作流程截然不同,但是您可以在了解它產生的立竿見影成效之後逐步採用這個嶄新的工作流程。

    透過本演講您將了解如何:
  • 使用Simulink進行Model based design 的優勢,通過更多實現細節進行協作來完善您的演算法
  • 建立可重複使用的測試平台基礎架構
  • 自動生成Verilog或VHDL,以及自動產生testbench驗證波形以加快RTL開發與驗證
  • 藉由演算法設計與數為硬體工程師之間的協作,能儘早匯整出高效率的硬體架構
  • 進行自動化和管理定點數的轉換過程,甚至利用原生浮點數實現到更有效的結果
  • 生成無錯誤的邏輯合成的RTL,實現到任何FPGA,ASIC或SoC的目標硬體
  • MATLAB搭配FPGA合成軟體可產生軟體(處理器)、硬體(FPGA)協同作業,讓設計SoC流程更容易

Michael Chen

TeraSoft Inc.

將Simulink模型共享為獨立應用程序,Web應用程序和企業應用程序

Share Simulink Simulations as Standalone Applications, Web Apps, and Enterprise Applications

摘要

Simulink建模已成為系統設計階段的關鍵部分,重用搭建好的模型,才能將其價值最大化,而這些作法都離不開模型的部署。Simulink Compiler為MATLAB的新工具箱,能夠將搭建完成的模型從設計階段擴展到部署階段,最大化的獲取模型的價值。

    通過Simulink Compiler,您可以:
  • 將模型部署為黑盒應用,進而保護你的知識產權;
  • 將模型用於桌面應用程序,或者伺服器應用程序;
  • 將模型與企業系統相集成,提供於非SimulinkR用戶使用;

本次演講,將向您介紹如果通過Simulink Compiler™,輕鬆地為各種場景部署Simulink模型。

Phoebe Li

TeraSoft Inc.

RF收發器設計

RF Transceiver Design

摘要

透過本演講,您將了解MATLAB和Simulink如何用來設計整合式RF收發器。從鏈路預算分析開始,考量阻抗失配(impedance mismatches)、非線性(non-linearity)、以及雜訊;使用量測元件的組合,比如S參數檔案,再搭配規格資料表,將可快速疊代你的設計規格,並且進行複雜的數位訊號處理演算法和控制邏輯整合。

演講焦點
  • 利用超外差(superhet)或直接架構轉換進行RF收發器的設計
  • 預算分析,包含非線性、雜訊、阻抗失配
  • 可適性RF收發器的模擬,包括數位訊號處理和控制邏輯

Learn how MATLAB and Simulink can be used to design integrated RF transceivers. Get started with link budget analysis, taking into account impedance mismatches, non-linearity, and noise. Use a combination of measured components, such as S-parameters files, along with data sheet specifications. Rapidly iterate on your design specifications and integrate complex digital signal processing algorithms and control logic.

Highlights
  • Design of RF transceivers using superhet or direct conversion architectures
  • Budget analysis including non-linearity, noise, impedance mismatches
  • Simulation of adaptive RF transceivers including digital signal processing and control logic

Giorgia Zucchelli

MathWorks Inc.

天線設計和RF傳播

Antenna Design and RF Propagation

摘要

設計天線和陣列天線不必然需要先成為EM專家。本演講將帶您從選擇天線規格開始,並快速地優化頻寬、指向性、以及其他設計中的重要因素;以及如何將天線安裝在一個大型架構中,並透過3D地形與城市圖視覺化呈現RF傳播效果。本演講也將展示不同傳播模型的使用,如Longley Rice或2次反彈光線追蹤(2 bounce ray tracing)等。

演講焦點
  • 使用矩估計(Method of Moments)進行天線的設計和分析
  • 大型架構的天線安裝
  • 於3D地形圖與城市圖上視覺化呈現RF傳播效果

Design antennas and arrays without the need to become an EM expert. Get started with selecting the antenna specifications and rapidly optimize bandwidth, directivity, and other critical aspects of your design. Install the antenna on a large structure and visualize RF propagation effects on 3D terrain and urban maps. Use different propagation models, such as Longley Rice or 2 bounce ray tracing.

Highlights
  • Design and analysis of antennas using the Method of Moments
  • Installation of antenna on large structures
  • Visualization of RF propagation effects on 3D terrain and urban maps

Giorgia Zucchelli

MathWorks Inc.

混合訊號設計與驗證

Mixed Signal Design and Verification

摘要

本演講將展示如何快速地設計和模擬混合訊號元件,例如PLL以及資料轉換器(Data Converters)。探索規格時,須將雜訊、非線性、時序缺陷(timing imperfections)等納入考量,接著使用測試平台來測量系統的重要性能表現,比如PLL相位雜訊的驗證。最後我們將展示如何與Cadence設計工具整合並匯入Spice網路連線表來進行由下而上的建模。

演講焦點
  • 設計與分析PLL和資料轉換器
  • 驗證相位雜訊、擾動、時序錯誤等設計缺陷
  • 透過與Cadence Virtuoso的整合及Spice網路連線清單匯入進行由下而上的建模

Rapidly design and simulate mixed-signal components such as PLL and Data Converters. Explore the specifications taking into account noise, non-linearity, and timing imperfections. Use testbenches to measure the critical performance of your system, for example to verify the phase noise of a PLL. Integrate with Cadence design tools and import Spice netlists for bottom-up modeling.

Highlights
  • Design and analysis of PLLs and Data Converters
  • Verification of design imperfections such as phase noise, jitter, timing errors
  • Bottom-up modeling through integration with Cadence Virtuoso and Spice netlists import

Giorgia Zucchelli

MathWorks Inc.

SerDes設計和IBIS-AMI模型產生

SerDes Design and IBIS-AMI Model Generation

摘要

本演講將使用FFE、CTLE、DFE等演算法為SerDes系統開發與分析發射器和接收器等化演算法的過程。於探索規格時,須考慮通道分散(channel dispersion)、串音(cross-talk)、和擾動(jitter)等,之後使用量測的資料以精準地建立通道模型或CTLE曲線;最後我們將展示如和自動產生雙重dual IBIS-AMI模型,並分享給同事和客戶。

演講焦點
  • 利用NRZ、PAM3、PAM4調變來進行SerDes系統等化器演算法的設計和分析
  • 驗證通道分散、串音、擾動帶來的影響
  • 自動為通道分析和回歸測試產生雙重IBIS-AMI模型

Develop and analyze transmitter and receiver equalization algorithms for SerDes systems using FFE, CTLE, and DFE algorithms. Explore the specifications taking into account channel dispersion, cross-talk, and jitter. Use measured data to accurately model the channel or CTLE curves. Automatically generate a dual IBIS-AMI model to share with your customers and colleagues.

Highlights
  • Design and analysis of equalization algorithms for SerDes systems using NRZ, PAM3, PAM4 modulations
  • Verification of the effects of channel dispersion, crosstalk, and jitter
  • Automatic generation of dual IBIS-AMI models for channel analysis and regression testing

Giorgia Zucchelli

MathWorks Inc.

利用MathWorks和Cadence工具設計和驗證混合訊號系統

Design and Verification of Mixed-Signal Systems with MathWorks and Cadence Tools

摘要

Scott Li

Cadence

自動駕駛輔助系統的即時原型化:車道維持與車道跟隨輔助系統

Real-Time Prototyping and Testing for ADAS: Lane Keeping and Following Assist Systems

摘要

車道維持輔助系統(LKA)和車道跟隨輔助系統(LFA)是高階自動駕駛輔助系統(ADAS)和自動駕駛(AD)的重要功能,它們會自動採取措施使車輛保持在車道內,並在安全距離內跟隨車輛,因此在設計可靠而強大的LKA和LFA系統時,必須考慮道路和駕駛條件的變化。

透過模擬駕駛來進行即時模擬和測試可以幫助驗證這些系統的設計,在本演講中,我們將討論如何透過在Simulink中所整合的虛幻引擎(Unreal Engine),利用逼真的模擬環境來模擬道路和其他條件,即時硬體系統我們所使用的是多核的Speedgoat,透過這樣的軟硬體環境來模擬車輛動力學和動力系統(powertrain)、以及車道維持和模型預測控制。以視覺為基礎的車道偵測演算法,是從Speedgoat內去實現Simulink可編程的FPGA I / O模組,因此可以處理來自Unreal Engine的影片資料數據。

Lane keeping assist (LKA) and lane following assist (LFA) systems are an important feature of advanced driver assistance systems (ADAS) and automated driving (AD). They automatically take action to keep the vehicle in its lane and follow cars at a safe distance. Variations in road and driving conditions must be considered when designing reliable and robust LKA and LFA systems.

Real-time simulation and testing can help verify these systems with simulated driving. We will discuss how to simulate road and other conditions using a photorealistic simulation environment leveraging the Unreal Engine from within SimulinkR. Using a multicore Speedgoat system, we simulate the vehicle dynamics and powertrain as well as lane keeping and following with model predictive controls. The vision-based lane detection algorithm is implemented on a Simulink programmable FPGA I/O module from Speedgoat, making it possible to process video data from the Unreal Engine.

Sarah Hung

TeraSoft Inc.

能元科技案例分享 - 電池物理模型的建立:使用Simulink和Simscape

Physical Modeling for Batteries:Using Simulink and Simscape

摘要

在這個演講中,能元科技與鈦思科技工程師將共同分享如何使用Curve Fitting與最佳化的函式,搭配電池充放電資料的蒐集,來獲取精準的Simscape電池模型參數。當準確的電池模型獲得之後,並利用其他的電池充放電資料來驗證模型的精準度。

Jerry Tung

TeraSoft Inc.

利用MATLB/Simulink來開發基於AUTOSAR的電池管理系統

AUTOSAR-Based Battery Management System Design by MATLAB/Simulink

摘要

AUTOSAR(汽車開放系統架構)是由許多汽車製造商、供應商共同開發之開放且標準化的汽車軟體架構,其分層軟體架構使得基於AUTOSAR標準設計之軟體,能夠應用於各種不同硬體及系統架構,因此適用各種日趨複雜的汽車平台,軟體開發者亦可專注於軟體應用層開發,包含:建模、模擬和驗證。

我們的電池管理系統軟體應用層採用由上而下(Top Down)開發流程,透過在AUTOSAR中建立系統架構與定義軟體組件描述,將組件描述(.ARXML文件)匯入Simulink,產生對應的Simulink SWC模型。接著,開發者在對應模塊中填入演算法: 如SoC電量估測,SoH健康度估測,控制邏輯,診斷等,以完成初始設計建模。

完成建模後,我們致力於模擬驗證設計以及達成自動化模型測試。利用Simulink Coder將模型轉換生成符合AUTOSAR標準的程式碼,並燒錄至微控制器以進行測試。透過SIL/ PIL/ HIL對電池管理演算法進行大量的模擬測試,從而顯著減少軟體問題。模擬電池物理模型所使用的電池模型參數則是對目標電池進行脈衝充放電實驗與等效模型擬合參數而得。

基於AUTOSAR與Maltab/Simulink模型的設計架構不僅能夠增加軟體模組的重複利用,也減少手工編碼,提高與客戶的溝通效率,以創造出高品質的電池管理系統。

林懿伶

ITRI, 機械所

鄭逸倫

ITRI, 機械所

不確定參數之最佳激發與校準:以車輛模型開發為例

摘要

陳怡平

國立台灣大學, 機械工程學系

PLCs與其它工業控制器的設計以及嵌入式演算法部署

Designing and Deploying Embedded Algorithms on PLCs and Other Industrial Controllers

摘要

在本演講,我們將展示工業系統工程師如何能夠在沒有實體的原型建造下,利用桌面電腦模擬來進行設計和測試控制邏輯和預測演算法。透過自動生成的C / C ++ 程式碼和符合IEC 61131-3標準的程式碼,您將可以加速嵌入式演算法部署到工業控制器(如PLCs)上,並保持硬體平台的獨立性。

我們展示如何利用基於模型的設計、利用工業系統的模擬模型來開發控制邏輯和狀態監控演算法,並自動產生PLC程式碼,以及執行實時測試的完整流程。

In this session, we will show how industrial systems engineers can use desktop simulation to design and test control logic and predictive algorithms without the need for a physical prototype.Through automatic generation of C/C++ code and code compliant with the IEC 61131-3 standard, you can accelerate deployment of embedded algorithms onto industrial controllers like PLCs, and stay hardware platform independent.

We show how to leverage simulation models of industrial systems using Model-Based Design to develop control logic and condition monitoring algorithms, automatically generate code for PLCs, and perform real-time testing.

Kary Chang

TeraSoft Inc.

MBD軟體設計與驗證於車用產品開發

摘要

利用MATLAB/ Simulink 實現 Model-Based Design (MBD) 的車用元件軟韌體開發。而車用開發須遵守標準流程 ASPICE。我們使用相關套件進行自動化檢測,以符合規範並提升軟體開發品質。

本題目將以實際產品專案為例,介紹MBD 軟韌體模型、虛擬研證平台、自動代碼產出(ACG) 以及軟韌體品質檢測方法。

蔡松霖

Delta Electronics

如何橋接系統工程的架構模型與模型化基礎設計

Bridging the Gap Between Systems Engineers' Architecture Models and Model-Based Design

摘要

系統工程師經常面對的問題是,如何在其所建立的系統模型與後續對應的軟體設計與實現之間,建立良好的整合。MathWorks全新發表的System Composer工具與MATLAB/Simulink之間可以構成一個高度整合的開發環境,在單一環境下一氣呵成的完成從系統工程、設計、實現與測試驗證的所有工作。在這個演講中,將說明系統工程師可以如何使用System Composer建立系統模型並與基於Simulink的Model Based Design環境緊密結合。演講重點將包括:

    焦點內容:
  • 在早期架構工作和下游設計之間架起一座橋樑
  • 創建架構模型並通過原型和配置文件擴展語言
  • 架構分析
  • 轉向設計和實施

Systems engineering is a challenging problem, and often the tools used to tackle these challenges do not connect well to the other tools used throughout the design process. MathWorks systems engineering tools combine with MATLABR and SimulinkR to create a unified modeling environment, enabling the use of a single platform throughout systems engineering, design, implementation, and verification processes tools.

In this talk, we present a workflow for systems engineering and architectural modeling with a tight connection to Model-Based Design.

  • Building a bridge between early architecture work and downstream design
  • Creating architecture models and extending the language through stereotypes and profiles
  • Analyzing architectures
  • Moving to design and implementation

Jerry Tung

TeraSoft Inc.

利用MATLB及Simulink進行工業機器人系統設計

Industrial Robotics System Design with MATLAB and Simulink

摘要

自主機器人技術需要許多工程領域的知識和經驗,包括機械設計、感知、決策、控制設計和嵌入式系統等。本演講將介紹完整的自主機器人之設計工作流程,使工程師可以輕鬆地學習和應用機器人的許多功能領域。我們將以機器人手臂拿取-放置(pick and place)為例,將逐步展示整個完成應用程序的開發。

    內容主題:
  • 開發機器人的運動學(kinematic)和動力學模型
  • 使用深度學習的感知演算法設計
  • 達成精確定位的多感測器融合做法
  • 用於感測器模型和環境模擬的Gazebo協同模擬
  • 避障(obstacle avoidance)路徑規劃
  • 使用Stateflow進行監督邏輯和控制

Autonomous robotics requires knowledge and experience in many engineering domains, including mechanical design, perception, decision making, control design, and embedded systems. This talk explains a complete autonomous robotics workflow that allows an engineer to easily learn and apply the many functional domains of robotics. We will walk through the development of a robot arm pick-and-place application.

    Some of the topics that will be covered include:
  • Developing kinematic and dynamic models of robots
  • Perception algorithm design using deep learning
  • Multi-sensor fusion for accurate localization
  • Gazebo cosimulation for sensor models and environment simulation
  • Path planning with obstacle avoidance
  • Supervisory logic and control using Stateflow®

Sarah Hung

TeraSoft Inc.