Abstract

This study presents a novel discrete-event systems (DES) modeling framework to address real-time system (RTS) with sporadic, periodic, and non-repetitive real-time tasks. Our approach is organized in three steps. First, the effect of individual timing parameters of each task, such as job arrival and deadlines, are represented by modular DES. Second, we choose the required modules for the specific RTS at hand to compose an overall model. Third, we utilize supervisory control to find all schedules that are consistent with the timing requirements of all tasks. In contrast to fixed task priorities, we address general preemption relations represented by a preemption matrix and thereby implement priority-free conditionally-preemptive (PFCP) real-time scheduling. As a particular feature of the closed-loop configuration, the schedules obtained refer to the actual job execution time as opposed to upper and lower bounds. We illustrate our approach by a real-world example in the context of an automated manufacturing system.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call