Hardware-Aware, Model-Based Software Development to Speed Up Embedded Designs

Overview

The complexity of Embedded Software Applications is significantly increasing, with the number of lines of code in today’s systems, such as cars, industrial machinery, airplanes, trains, and medical devices, in the order of millions. This is driving the need for embedded software engineers to work at higher levels of abstraction and to move to model-centric development workflows, such as Model-Based Design, defined as ‘The systematic use of models throughout the development process’.

Such an approach allows engineers to model software algorithms together with the environment they are designed to interact with, validate them through simulation and generate the corresponding embedded code. This workflow provides several advantages, as it permits to shorten the development time by shifting-left verification and validation, automatically generating embedded code and abstracting from the hardware architecture. If kept fully hardware-agnostic, this would lead to sub-optimal and incomplete code, since the hardware-specific aspects would not be fully exploited and integrated. Two typical examples of such aspects, playing a key role in software optimization and integration, are device drivers and hardware accelerators. In this online session we will show how to overcome these drawbacks through a hardware-aware, model-based software development workflow.

Highlights

• Typical and Model-Based Design workflows for embedded systems
• Integration of platform software, drivers, and libraries
• Hardware-aware applications modeling
• Generation of C code from models
• Simulation based Verification and Validation
• On target applications deployment and data visualization

About the Presenters

Mauro Fusco is an Application Engineer at MathWorks in Eindhoven. He specializes in supporting customers in Aerospace, Automotive and Machinery industries for the establishment of Design Automation workflows. Modelling, simulation, testing and implementation through automatic code generation whilst conforming to international standards are key aspects of his work.

Before joining MathWorks, he worked for 5 years at the Dutch Organization for Applied Research, TNO, focusing on the domain of Controls for Cooperative and Autonomous Driving. Mauro has a Masters in Automation Engineering from the University of Naples Federico II, during which time he conducted research at Eindhoven University of Technology. His technical expertise lies in the areas of Control Theory, Non-linear and Network Control and their implementation.

Irina Costachescu is a Software Engineer with the Model-Based Design Team at NXP Semiconductors in Bucharest, Romania. She holds a Bachelor’s Degree in Automatic Control and Systems Engineering and a Master’s Degree in Complex Systems, both from the Polytechnical University of Bucharest. During her studies Irina deepened her control systems theory knowledge by developing robust and optimal algorithms for various systems control while also being a teaching assistant for this subject. At NXP, Irina contributes to all the development phases, from framework design to peripherals support, working collaboratively with internal teams for integrating other tools into the Model-Based Design Toolbox, dedicating her time also to creating toolbox related webinars, videos, and training sessions.