View article

The efficient development of embedded controllers for Cyber-Physical Systems (CPS) is under unprecedented pressure, driven by an upsurge of markets combining the Internet of Things (IoT) and flexible production systems (Marwedel 2018). Salient properties in this setting are the uncertainty and rapid evolution of requirements, pushed by a pervasive availability of cheap, yet powerful technology for sensors, actuators, and embedded computing platforms. For decades the software engineering community has spent a tremendous effort in creating formal methods and tools to develop controllers for embedded systems, in particular for those of hybrid nature and with real‐time constraints. Modern design methods for hybrid controllers tend to rely on unified modeling frameworks, which capture and combine together the expressive power of well‐-known modeling techniques such as hybrid automata (see, for instance, Branicky et al.(1998) and references therein).

Yet, most existing methods are still heavy, expensive, and hard to scale up for real applications. Model‐and simulation‐based techniques (Jensen et al. 2011) offer increasingly attractive capabilities to allow for rapid, yet robust prototyping and final delivery of embedded systems (Wainer and Castro 2011). Recently in (Tolk et al. 2018) it was recognized that Model‐Driven Engineering is a strong