Evolution of an Automotive Modelling Language for Enhanced Support of Diverse Network Interface Controllers

Over the last two decades, vehicles have undergone a significant shift, transforming into highly software-intensive systems. Projections indicate that even entry-level vehicles will soon integrate hundreds of millions of lines of code and incorporate numerous electronic control units. To navigate the complexity of these software-intensive systems, there has been a notable shift towards adopting model-driven engineering and specialised modelling languages. Among these languages, the Rubus Component Model has played a crucial role for over 25 years, supporting the development and timing analysis of distributed resource-constrained embedded systems. The enduring success of the Rubus Component Model lies in its responsiveness to end-users’ demands and its ability to adapt to technological advancements. Notably, the proliferation of network interface controllers, including controller area network controllers, supporting diverse message-receiving policies like polling and interrupt, represents a significant advancement. However, the implications of these policies on end-to-end delays in distributed systems necessitate explicit modelling and dedicated timing analysis tools.This paper introduces an evolved Rubus Component Model, tailored for model-driven development and timing analysis in distributed embedded systems that utilise network interface controllers with diverse message-receiving policies. Drawing inspiration from a real-world example, the paper introduces new elements and properties in the Rubus Component Model designed to support these policies and facilitate timing analysis. The practical application of these enhancements is demonstrated, and insights are extended to other contemporary modelling languages in the vehicular domain. Beyond bolstering expressiveness, this evolution ensures the timing predictability of distributed embedded systems, aligning seamlessly with the Rubus Component Model’s core focus.