Introduction to the Special Issue on Time for CPS (TCPS)

Abstract

For many Cyber-Physical Systems (CPS), timing is crucial for safety, security, and responsiveness of the system behavior. Time is key to enabling coordinated actions among the many, often heavily distributed, components of a CPS. For example, in power systems, the time of all phasor measurement units (PMUs) is synchronized via GPS signals, because otherwise aligning data from various distributed PMUs will become impossible, rendering state estimates wrong and unusable. With the increasing connectivity in modern CPS, requirements on timing accuracy and synchronization are evolving, ranging from tight, picosecond synchronization accuracy in power systems to high precision and accuracy requirements for wireless and low-power networks. Smart cities and connected vehicles pose new technological challenges and timing properties play an important role for coordination and security. Despite the importance of time in CPS, there are significant gaps in specifying, reasoning about, verifying, and testing the timing behavior of systems. In practice, timing in CPS is often an afterthought in the development process. While experienced domain experts might understand the desired timing behavior of the CPS, they often do not have a standardized, formal way of describing the timing requirements, let alone incorporating timing properties as part of the design. Even if a design is accompanied with well-defined timing requirements, it is difficult to verify whether a given design satisfies those requirements. The article in this special issue address challenges ranging from specifying, modeling, and verifying time in CPS in various application domains, including automotive control, communication, and manufacturing. • In their work on “Composable Finite State Machine–based Modeling for Quality-ofInformation-Aware Cyber-physical Systems,” Rafael Rosales and Michael Paulitsch present a model-based design methodology and introduce composable design patterns to address the following Quality-of-Information properties: timeliness, correctness, completeness, consistency, and accuracy. By specifying and composing behaviors using extended finite state machines, reuse and robustness are increased and dynamic validation and optimization of functional and nonfunctional properties is enabled. • The article “System-level Logical Execution Time: Augmenting the Logical Execution Time Paradigm for Distributed Real-time Automotive Software,” by Kai-Björn Gemlau, Leonie Köhler, Rolf Ernst, and Sophie Quinton, apply the well-known logical execution time paradigm, which abstracts away notoriously hard-to-characterize and often non-deterministic physical execution times, not just to a single component but also in a systemwide context. By explicitly acknowledging the fact that communication times are not negligible and cannot be abstracted way, the work addresses challenges in the design and verification of complex automotive systems, such as predictability, synchronization, composability, and extensibility.