• Successfully apply separation for mixed criticality
• Implications of soft vs. hard real-time
• Tailoring a system architecture for safety & security
• Different markets/applications—same requirements

• Avoid coding pitfalls, find remaining bugs faster
• Understand any system through full in-system visibility
• Easier debugging of multicore systems
• Efficiently collaborate across distributed teams

Embedded systems are becoming more complex, requiring higher performance and more features, leading to an increased demand for modern programming languages such as C++. In this class, we will discuss the use of C++ and modern C++ (C++11,C++14, etc.) for embedded development, focusing on its benefits and challenges. Attendees will learn: C++ and Modern C++ benefits, challenges, best practices for embedded development, as well as Modern C++ constructs.

In the world of embedded systems, low-latency performance is crucial for ensuring speed, reliability, and responsiveness. This presentation delves into the core strategies for optimizing embedded applications to minimize latency, highlighting why low-latency is critical in sectors such as autonomous vehicles, industrial automation, and medical devices.

We explore the key factors contributing to latency at both the hardware and software levels, including processor limitations, communication bus delays, and task scheduling inefficiencies. Techniques to reduce latency are discussed, using low-level programming techniques, and adopting hardware-software co-design methodologies. Through real-world examples and case studies, the presentation demonstrates how these techniques can drastically enhance system performance and ensure real-time operations.

Modern System-on-Chip (SoC) architectures are characterized by the integration of multiple CPU cores, hardware accelerators, and peripheral controllers, all designed to deliver optimized performance for a wide range of applications. This level of integration introduces significant complexity in the development and management of embedded software, and engineers must navigate intricate dependencies, optimize resource utilisation, and ensure system reliability during development. This paper examines the critical obstacles posed by such architectures, focusing on the difficulty of managing hardware abstraction, synchronisation between software modules, and cross-platform compatibility. Additionally, it highlights key opportunities such as the use of middleware frameworks, hardware-agnostic development practices, and enhanced development tools that can streamline the software creation process. By addressing both the challenges and opportunities, this work aims to offer practical insights for embedded software engineers seeking to optimize their workflows and harness the full potential of complex SoC devices.

Agile methods are by definition in contradiction to the static development model of a typical software project that has to be certified according to functional safety standards. Safety standards, such as IEC 61508 or ISO 26262, require a development process based on a V-model. This also means, if the top-level requirements change, the entire V-model cycle must be completed before a new release can be created. This approach does not fit with the Agile Manifesto, which advocates, for example, flexible responses to changing requirements and regular releases. Nevertheless, there are ways to develop safety projects using agile methods. This contribution is a field report based on two safety projects in which agile methods were applied at specific parts of the process. It explains which aspects of agile development fit at which points in the V-model and which do not, and which lessons were learned when applying them.