2015 MATLAB® & Simulink® 通訊、影像視覺及馬達控制技術研討會




議程說明

MATLAB & Simulink R2016a 版本於物聯網應用之最新功能及工作流程
(Highlights of MATLAB & Simulink 2016a for IoT Workflows)

物聯網(IoT, Internet of Things)描繪了一股新興的大趨勢,代表著大量的嵌入式設備(物)都被連接至網際網路;而這些被連接的設備與人、其他設備進行溝通後,透過感測器將資料儲存雲端上,藉由其上的運算資源進行資料處理及分析,而獲得重要的資訊及洞見。而雲端運算能力的日益平民化讓人們負擔得起,再加上設備間的連接性愈來愈強,因此推波助瀾助長了這股趨勢。IoT系統的成功必須仰賴健全的資料解析能力,以及可以在嵌入式設備、雲端、或離線的桌上型電腦上進行設計的強大技術能力。

IoT解決方案是因應許多垂直的應用而產生的,如環境監測及控制、健康監測、車隊監控、工業監測及控制與家用自動化等等。MATLAB® and Simulink® 可藉由幫助您開發及測試邊緣節點設備(edge node devices)、取得及匯集資料、分析IoT感應資料及建立傳輸通道模型等工作。

在本演講中,我們將先介紹典型的IoT工作流程及其挑戰。之後的幾個演講,將依照本工作流程依次深入探討MATLAB與Simulink如何助您應對這些挑戰。

Internet of Things (IoT) describes an emerging trend where a large number of embedded devices (things) are connected to the Internet. These connected devices communicate with people and other things and often provide sensor data to cloud storage and cloud computing resources where the data is processed and analyzed to gain important insights. Affordable cloud computing power and increased device connectivity is enabling this trend. IoT systems depend upon robust data analytics, the design for which can occur on embedded devices, in the Cloud or offline on the desktop.

IoT solutions are built for many vertical applications such as environmental monitoring and control, health monitoring, vehicle fleet monitoring, industrial monitoring and control, and home automation. MATLAB® and Simulink® products support IoT systems by helping you develop and test edge node devices, access and aggregate data, and analyze IoT sensor data and model communications channels.

In this overview session, we will describe the typical IoT workflow and its challenges. We will come back to this workflow in later sections of the seminar as we dive deeper into how MATLAB and Simulink can help address these challenges.


資料解析的方法介紹
(Introduction to Data Analytics)

工程資料(Engineering data)在企業關鍵系統(business-critical systems)及應用變得愈來愈重要。聲音、影像、即時視頻、動作、機器性能的公制單位、以及其他感測器產生的資料,未來將會與企業、交易以及其他的IT資料結合,以對複雜的現象進行更精密的分析。除此之外,資料的大小也將帶來許多不同的挑戰;比方說,記憶體不足、處理時間過長、或資料產生太快來不及儲存等等。標準的演算法通常不是為了在合理的時間及記憶體空間下處理龐大資料集而設計的。

在本演講中,您將學會利用MATLAB®及技巧來處理這些挑戰。本演講將聚焦於一般性的方法,接下來的演講中,將對特定的工程資料類型進行詳細的介紹。

In this session you will learn approaches and techniques available in MATLAB® to tackle these challenges. We will focus on general approaches in this talk and specialize into specific types of engineering data in later sections.

Engineering data have become essential in business-critical systems and applications. Audio, image, real-time video, motion, machine performance metrics, and other sensor-generated data are being combined with business, transactional, and other IT data to create opportunities for sophisticated analytics on more complex phenomena. In addition, the size of the data sets may present challenges such as lack of available memory, taking too long to process, or streaming too quickly to store. Standard algorithms are usually not designed to process big data sets in reasonable amounts of time or memory.


感測器解析 Part 1 透過資料採礦洞見分析目標及獲得重要知識
(Gaining Insights into Activities through Data Mining)

有愈來愈多的應用需要在時間序列及感測器資料上,同時進行訊號處理及機器學習的處理技巧。MATLAB可以在單一的環境下提供一系列完整的建模及設計能力,幫助加快資料的解析以及感測器處理系統的開發。

本演講我們會介紹MATLAB常見的訊號處理方法(包含數位濾波器及頻域分析),有助於從原始波形萃取出可描述的特徵,我們也將為您展示如何利用平行運算來加速大量資料集的處理。接下來,我們會討論如何利用程式及互動的方式來探索、測試不同類別的演算法(例如決策樹、支持向量機、或類神經網路)。

我們也將演示如何利用MATLAB轉檔佈署的工具,來建構一個簡潔流暢的嵌入式感測器分析的分類演算法。

An increasing number of applications require the joint use of signal processing and machine learning techniques on time series and sensor data. MATLAB can accelerate the development of data analytics and sensor processing systems by providing a full range of modelling and design capabilities within a single environment.

In this session, we introduce common signal processing methods in MATLAB (including digital filtering and frequency-domain analysis) that help extract descripting features from raw waveforms, and we show how parallel computing can accelerate the processing of large datasets. We then discuss how to explore and test different classification algorithms (such as decision trees, support vector machines, or neural networks) both programmatically and interactively.

We also demonstrate the use of MATLAB deployment tools to architect a streaming classification algorithm for embedded sensor analytics.


感測器解析 Part 2 設計資料精簡 (Data Reduction)的電腦視覺演算法
(Designing Computer Vision Algorithms for Data Reduction)

近年來由於電腦視覺演算法朝強健以及高效率發展,以攝影機為基礎的系統已成為高級輔助駕駛系統(ADAS, advanced drive assistance systems)主要的途徑。這些視覺系統快速地變成主動式安全系統的重要構成元件。

在本演講中,我們將展示如何進行邊緣設備的分析並精簡資料的維度,以輕鬆地在雲端匯集資料。尤其,我們將說明MATLAB®如何被運用在電腦視覺化的基礎分析及ADAS元件的設計,像是以電腦視覺演算法進行行人偵測,以及以立體視覺(stereo-vision)為基礎的防撞設計等。除此之外,我們還會展示如何針對開發的演算法自動產生程式碼,然後在嵌入式的感測器平台上運行程式碼。

Camera-based systems have become a key approach to advanced driver assistance systems (ADAS) due to recent advancements in robust and efficient computer vision algorithms. These vision systems are rapidly becoming a key component of active safety systems.

In this session, we show how to develop analytics and reduce data dimensionality at the edge node for easier aggregation in the Cloud. Specifically, we demonstrate how MATLAB® can be used to design computer vision-based analytics and ADAS components such as pedestrian detectors and stereo-vision-based collision avoidance using computer vision algorithms. Additionally, we will show how to automatically generate code for the developed algorithms and run them on an embedded sensor platform.


在ARM及FPGA為基礎的邊緣設備上實現IoT演算法
(Implementing IoT algorithms onto ARM- and FPGA-based edge devices)

Simulink支援邊緣節點硬體平台的原型化設計以建立物聯網系統,您可以在Simulink中開發預處理的演算法,然後使用HDL和C程式碼產生的產品進行轉檔,並佈署到嵌入式硬體上。轉檔部署演算法到設備的運算,藉由這些裝置計算能力的優勢,大幅降低了必須發送給雲端數據聚集器的數據量。

在本演講上,我們將展示如何使用Simulink硬體支援包(Simulink hardware support package)進行演算法硬體原型化,以及如何將MATLAB演算法產生轉成嵌入式C程式碼。

本演講可協助正使用MATLAB開發和測試MATLAB演算法的工程師,幫助他們將演算法放到邊緣設備上。

Simulink supports edge node hardware platforms for prototyping and building IoT systems. You can develop preprocessing algorithms in Simulink and then deploy them on your embedded hardware using HDL and C code generation. Deploying preprocessing algorithms on the devices' computing capabilities takes advantage of the computing capabilities on these device and reduces the amount of data that must be sent to the cloud data aggregator.

In this session, we show how to prototype your algorithms with Simulink hardware support package, as well as to generate embedded C code from your MATLAB algorithms.

This session is geared toward algorithm engineers developing and testing algorithms in MATLAB who are looking to put algorithms on edge devices.


資料匯集並將IoT解決方案配置到雲端上: 資料解析與企業系統的整合
(Aggregating Data and Deploying IoT Solutions to the Cloud: Integrating Analytics with Enterprise Systems)

使用資料解析方法將大量的複雜資料轉換成可以執行的資料可以幫助您改善設計及決策流程。然而,開發有效的解析方法並將整合至企業商業系統是非常具挑戰性的。在本演講,您將學習如何利用MATLAB®提供的方法及技巧來處理這些難題。

若要與其他人分享MATLAB程式,MATLAB演算法可以與伺服器為基礎的網路及企業系統做整合,因此提供了最大程度的規模拓展度、客製化、且可與企業IT架構整合。以MATLAB為基礎的程式不僅可以轉檔配置為獨立的應用程式,也可以作為軟體元件被整合到網路及企業應用程式中。

我們將展示MATLAB這些新功能的設計及開發,讓您成為這個以資料解析為主流的最新世代的領導者。

Using Data Analytics to turn large volumes of complex data into actionable information can help you improve design and decision-making processes. However, developing effective analytics and integrating them into business systems can be challenging. In this session, you will learn approaches and techniques available in MATLAB® to address these challenges.

To share MATLAB programs with others, you can integrate MATLAB algorithms with server-based web and enterprise systems, providing maximum scalability, customizability, and integration with your organization’s IT infrastructure. Your MATLAB based programs can be deployed as standalone applications and also as software components that can be integrated into web and enterprise applications.

We will demonstrate the design and development of these new capabilities in MATLAB, which will empower you to be a leading force in this new analytics-driven age.