How to Build an Autonomous Anything


Autonomous technology will touch nearly every part of our lives, changing the products we build and the way we do business. It’s not just in self-driving cars, robots, and drones; it’s in applications like predictive engine maintenance, automated trading, and medical image interpretation. Autonomy —the ability of a system to learn to operate independently—requires three elements:

  • Massive amounts of data and computing power
  • A diverse set of algorithms, from communications and controls to vision and deep learning
  • The flexibility to leverage both cloud and embedded devices to deploy the autonomous technology

In this talk, Richard shows you how engineers and scientists are combining these elements, using MATLAB® and Simulink®, to build autonomous technology into their products and services today—to build their autonomous anything.

Richard Rovner, MathWorks

Two Worlds Coincide: Financial Risk Management and Model-Based Design


At first glance, financial risk management might appear to have little in common with engineering Model-Based Design. Financial risk management is data-centric, highly dimensional, and deployed into software systems. Engineering models typically draw on fewer, highly coupled inputs, often embedded in physical and electronic hardware.

In both cases, validated, verified fit-for-purpose models are key, extending product lifecycles amidst extreme scenarios, albeit across different time horizons. Good processes mitigate against risks such as costly trading errors or compliance charges in finance, while high integrity requirements have long dominated engineering. Fit-for-purpose models also increase functionality and drive progress, enabling more differentiating features on a car, device, or plane, and facilitating new investment, lending, and liquidity-creating products.

In this talk, Ray discusses how financial risk technology stacks are evolving in response to regulatory and geopolitical change, bigger datasets, new modelling techniques, and rapidly changing development cultures. He also assesses the critical importance of good model development and implementation, and what insights he has taken from Model-Based Design in other industries.

What’s New in MATLAB R2017a and R2017b


In this session, Ned Gulley introduces new capabilities in the MATLAB® product family in Releases 2017a and 2017b. Ned shares his insights into how MATLAB is designed to be the language of choice for millions of engineers and scientists worldwide.

Attend this session for a unique opportunity to learn from one of the key designers of MATLAB.

Ned Gulley, MathWorks

Big Data and Machine Learning for Predictive Maintenance


Predictive maintenance—the practice of forecasting equipment failures before they occur—is a high priority for many organisations looking to make critical engineering decisions from data.

Learn about new capabilities in MATLAB® for big data, machine learning, and deep learning, and how these can be combined with model-driven approaches to create and deploy predictive maintenance algorithms to embedded devices and cloud analytics platforms.

Paul Peeling, MathWorks

MatConvNet: Deep Learning Research in MATLAB


In this talk, Andrea will give an overview of the artificial intelligence and computer vision research being done by the Oxford Visual Geometry Group. He will show examples of object recognition, counting, text spotting, face verification, and many others. He will also demonstrate how much of their research work is done using Parallel Computing Toolbox™ on a large cluster of GPU-equipped computation servers. The University of Oxford builds and maintains their MATLAB® based deep learning library, MatConvNet. A spiritual successor to their earlier VLFeat project, this toolbox has been widely used in the computer vision community for almost a decade.

Andrea Vedaldi, University of Oxford

Audio Source Separation: "Demixing" for Production


"Live at the Hollywood Bowl" was the only live album from The Beatles, and a hard one to listen to because of the screaming fans drowning out the music. The simple concert recording technology available in 1964 provided only a 3-track recording that was impossible to remix effectively using traditional methods. However, Abbey Road Studios recently used MATLAB® to create a new proprietary algorithm that can separate out the individual original sources from the old recordings in order to remaster the recording and remove 95% of the background noise from the fans. In this talk, you will learn how a single engineer built a tool that exceeded any solution available on the market through community collaboration, code reuse, and the ability to quickly turn mathematical algorithms into an effective application. The complex tool they developed includes a mixture of spectral analysis, optimization, control theory, and musical modelling, and allowed them to release a new version of the album where the Fab Four can be heard live and like never before.

James Clarke, Abbey Road Studios

Reusing and Prototyping Code to Accelerate Innovation: Smart Voice Interfaces


Smart devices like the Amazon Echo are shifting expectations on the near future of human-machine interactions. As innovative products include more sensors and aim to deliver increasingly complex features, successful manufacturers and solution providers need to reuse more software assets and prototype more quickly than ever before.

In this session, you will learn about different MATLAB® capabilities that enable fast code integration, from embedded IP to cloud-based services. You will also discover techniques for very early prototyping, including methods for running in real time and connecting to real-world signals. The talk focuses on applications of signal processing and data analytics. As a worked example, we discuss the design of an IoT-enabled voice-driven interface.

Gabriele Bunkheila, MathWorks

Detection of the Security Feature in the New £1 Coin


In this session, you will learn how The Royal Mint used MATLAB® to develop the most secure £1 coin ever made. The new £1 coin incorporates advanced security measures, one of which is the inclusion of a chemical marker in the metal inner of the coin. Rhys Thomas, Sarah Rogers, and a small team of engineers at The Royal Mint leveraged image and signal processing functionality in MATLAB to develop algorithms capable of detecting the presence of the marker in genuine coins.

Rhys Thomas and Rogers will discuss the development of the new coin, the authentication algorithm, and how the team managed the large amounts of data generated by the project.

Ellis Rhys Thomas, The Royal Mint

Sarah Rogers, The Royal Mint

Scaling MATLAB for Your Organisation and Beyond


Do you need to work with ever-growing data volumes and complexity in your algorithms? Do you need to support an ever-growing number of users, who want their results faster?

In this session, Rory introduces the technologies that enable you to achieve your scaling goals, including:

  • Managing algorithm complexity and data volumes using additional hardware such as clusters and cloud computing
  • Deploying your algorithms and applications to a growing user base both inside and outside of your organisation

Rory Adams, MathWorks

Model‐Based Design for Fuel System Development


In modern civil aircraft, the fuel system provides more functionality than simply keeping the engines fed. It is also used to provide structural integrity of the wings and fuselage, to protect the aircraft flight envelope with centre‐of‐gravity control, to provide cooling for the engines and hydraulic packs, and to improve aircraft efficiency by trimming the aircraft in flight. These extra functional requirements, together with a plethora of safety requirements, can lead to complexities within the design of the fuel control system. To address them, Airbus has employed Model‐Based Design, including the use of MATLAB®, Simulink®, Stateflow®, and Simscape™. This presentation focuses on how system level simulation, comprising of controls models in Stateflow and hydraulics models in Simscape, is being used to support the Airbus Fuel Control System design. The advantages of Model‐Based Design to facilitate full validation and verification at the system and aircraft level are also discussed.

What's New in Simulink R2017a and R2017b


Recent updates to the Simulink® product family will help you maximise the efficiency of your implementation of Model-Based Design. Discover new capabilities in plant modelling, control design, and production code generation. Speed up simulations and learn about new productivity improvement techniques. Streamline your verification and validation tasks with enhanced automation functionality.

Join this session to get the most out of Model-Based Design with the latest features in Simulink.

Kate Thorne, MathWorks

Certifiable Production Code Development


High integrity software is now common place within many industries. As software complexity and programme demands increase, the level of quality cannot be compromised. This presentation discusses how Rolls-Royce uses Model-Based Design to address these issues for airborne software within their family of Engine Monitoring Units. By using a real-world example, the use of the tools offered by MathWorks in aiding the certification objectives of RTCA DO-178C are demonstrated from high level requirement to validated object code.

David Owens, Rolls-Royce

Using Simscape to Model a Formula E Powertrain


The Jaguar I-TYPE is Jaguar Land Rover’s Formula E car developed and operated in collaboration with Williams Advanced Engineering. This all-electric race car delivers 200 kW (270 bhp) to reach 0–60 mph in 2.9 seconds and a top speed of 140 mph. A successful race strategy depends not only on peak power output, but also on careful energy management to ensure sufficient range. In this talk, Dr. David Hinchley describes how Jaguar Land Rover modelled the Formula E powertrain using Simscape™, validated the model against race and test data, and undertook simulation studies to inform their energy management strategy.

Dr. David Hinchley, Jaguar Land Rover

How Simscape Supports Design Innovation for Cyber-Physical Systems


Do you need to model physical systems to support your development process? Do you compromise on design choices because you lack the data, tools, or the time to model them adequately? In this session, you will learn how Simscape™ helps engineers build models that are fit for purpose, supporting both early design decisions and subsequent integration and system test.

You will learn how Simscape can be used to model physical systems, starting with an overview of off-the-shelf Simscape libraries and building up to how you can create your own custom component models. The presentation is illustrated using the example of modelling electric drives for vehicle and robotic applications, including thermal and cooling aspects. Key themes include:

  • How to model at an appropriate level of fidelity
  • Ways to deal with lack of data (e.g. for supplier components)
  • Deploying models through code generation (e.g., to support hardware-in-the-loop testing)

Solutions to these themes exploit the tight integration between Simscape and MATLAB® and Simulink®, with MATLAB providing the numerical platform to manage your data and Simulink providing seamless integration with your control algorithms and path to real-time deployment.

Rick Hyde, MathWorks

Hardware Software Co-Design and Testing Using Simulink® Real-Time™


Real-time hardware-in-the-loop (HIL) models have been an integral part of system testing and acceptance for over 20 years. With the increased use of Model-Based Design throughout the algorithm development cycle and improved performance of real-time simulators, the HIL model becomes ever closer to the Model-Based Design model. This allows for improved traceability from the algorithm specification to the hardware implementation and associated test environment.

Brian Steenson, Thales

Paul Berry, Thales

Hardware and Software Codesign for Motor Control Applications


Electric motors are everywhere, and are finding new applications every day. The technology to control motors is also evolving to target new platforms comprising both hardware (FPGA/ASIC) and software (microprocessor), such as Xilinx® Zynq®. In this session, we looks at some of the challenges and solutions for developing motor control algorithms, using Model-Based Design, including:

  • Techniques to model and parameterise electrical machines for simulation work
  • Approaches to partition controller designs between hardware and software
  • Methods to generate code for the software, hardware, and the interfaces between them

GianCarlo Pacitti, MathWorks

Introduction to MATLAB


In this session, Lianne introduces MATLAB®, the interactive environment, and high-level language for numerical computation, visualisation, and programming. Topics discussed and demonstrated in this session include:

  • Importing data from Microsoft® Excel®, text files, databases, and devices
  • Exploring and visualising data using interactive tools
  • Performing mathematical analysis on the data
  • Automating your analysis and creating reports

See how easy it is to get started using MATLAB.

Lianne Crooks, MathWorks

Introduction to Signal Processing


Signal processing is essential for a wide range of applications, through communications, fault detection, and data science. In this session, Steven introduces how you can use MATLAB® for signal analysis and processing tasks across applications and workflows through examples covering:

  • Exploring signals to make measurements, perform spectral analysis, and design filters
  • Identifying features such as changes and sequences in signals
  • Analysing vibration measurements to understand vibrational modes
  • Integrating algorithms into system models

Steven Thomsett, MathWorks

Introduction to Simulink and Stateflow


In this session, Jonathan introduces the Simulink® product family. Topics include:

  • Basic concepts for Simulink and Stateflow®
  • Problems that are appropriate for Simulink rather than MATLAB®
  • Using Simulink and Stateflow to effectively capture your complete system modelling needs
  • How you can use code generation to take your designs from desktop to hardware

This presentation is ideal for Simulink beginners and MATLAB users interested in learning more about Simulink.

Jonathan Agg, MathWorks

Introduction to Machine Learning and Deep Learning


Machine learning is ubiquitous. From deep learning in medical diagnosis, speech, and object recognition to data-driven modelling for engine health monitoring and predictive maintenance, machine learning techniques are increasingly at the heart of today’s autonomous systems. In this session, Conor discusses both machine learning and deep learning techniques in MATLAB®, as well as addresses the computer vision problem of object detection and recognition. This introduction provides participants with an overview of the available techniques and how to get started with machine learning in MATLAB.

Conor Daly, MathWorks

Simulation, Prototyping, and Verification of Standards-Based Wireless Communications


MATLAB® and Simulink® products enable you to design and prototype wireless communications systems. Using these tools, you can go from algorithm design and simulation all the way through to code generation and prototyping on hardware platforms.

In this master class, we show how you can simulate wireless standards such as LTE, WLAN, and 5G candidate technologies, using SDR platforms to interface with real-life signals.

During this presentation, you’ll see how you can use these algorithms in models suitable for C and HDL code generation. Finally, you’ll learn how this generated code can then be targeted to a Xilinx® Zynq® SoC platform for rapid prototyping and verification.

Colin McGuire, MathWorks

Neil MacEwen, MathWorks

Toolbox Development


MATLAB® toolboxes are becoming the standard mechanism for creating and sharing collections of MATLAB files, including code, data, apps, examples, and documentation.

This master class answers questions and discusses challenges that teams and enterprises face when adopting MATLAB toolboxes, including how to create a consistent set of content, compatibility with MATLAB data type and other toolboxes, choosing when to create an app, and how to work with version control.

Amy Koh, MathWorks

David Sampson, MathWorks

Testing Simulink Models


Simulink® Test™ is designed to ease the authoring, managing, and reporting of tests, and is the key to enabling efficient Model-Based Design. In this session, Fraser shows examples of how you can:

  • Create and manage test harnesses for temporal and logic-based tests
  • Create different types of tests, including regression, equivalence, and requirement-based tests
  • Link tests and means of verification to requirements
  • Integrate custom verification criteria
  • Integrate formal analysis methods
  • Automate tasks using programmatic interfaces, including continuous integration and running tests for modified components of your design

Fraser Macmillen, MathWorks

Team-Based Collaboration in Simulink


Most engineers work in teams, requiring tools that support the creation of a shared team environment and the ability to partition large designs into manageable components. In this session, Sonia demonstrates the latest project management capabilities in MATLAB® and Simulink® that address these challenges and promote more efficient teamwork. Topics include:

  • Sharing work across a team
  • Componentisation
  • Peer reviews of software changes
  • Automation and reporting
  • Impact analysis to understand the implications of a change

Sonia Bridge, MathWorks