Towards real-time cyber-physical systems instrumentation for creating digital twins

One of the challenges with digital twins is data collection from the physical systems. Data collection for digital twins in Cyber-Physical Systems, where the controllers are often hard real-time, is a challenging task, as the functional system behavior may be impacted by the instrumentation. An often applied solution is the use of additional monitoring controllers, which passively monitor the system. However, this solution introduces additional points of failure, and is costly as more hardware is required. Therefore, in this paper, we propose an approach that aids the digital twin developer by finding the optimal instrumentation rate for a set of parameters that are to be instrumented on a hard real-time controller. The optimization ensures that the hard real-time constraints are guaranteed, while also taking into account the developer's preferences with regards to the instrumented parameters. We apply this process on two cases: an exploratory example and an adaptive cruise controller.