Advanced Methodologies for Real-Time Discrete Event Modelling and Simulation

实时嵌入式系统被构建为与其周围环境(交通、自动化建筑等)交互的组件集。它们的正确性以及它们的及时性都是至关重要的：执行系统任务的时刻可能会导致灾难性的后果(对物品或生命)。我们提出了在实时嵌入式环境中使用建模与仿真的进一步研究。我们建议使用形式化的建模和仿真方法，因为用于实时建模和仿真的软件构件质量较高，所以形式化方法可以提供良好的结果。特别是，我们专注于DEVS形式主义的应用，这是一个基于通用动态系统概念的形式化建模和仿真框架。我们建议将DEVS用于实时系统，使用DEVS理论。尽管DEVS建模的最新进展一直在稳步发展，但现有的应用于实时环境的DEVS研究成果中没有一个考虑到实时应用中的各种复杂问题：容错、传感器复制、瞬时系统过载等。我们将采用新的技术来解决这个问题，基于使用不精确的计算和调度方法，提出在瞬时过载下控制降级的新方法(通过将任务划分为强制部分以提供基本结果，并将可选部分划分为可选部分以改进所获得的结果；还可以提供同一任务的多个版本，根据当前负载使用不同的版本)。我们的研究计划的目标是结合各种理论框架，为基于DEVS的非精确实时计算提供一种技术。从方法论的角度来看，我们感兴趣的是发展一种使用非精确计算的动态RT DEVS理论。其目标是提供安全性，减少开发时间，并提供开发模拟的正式框架的优势。仿真结果可应用于实时系统的开发。从实际的角度来看，我们希望提供一套工具，可以用来开发实时软件和半实物仿真(需要实时响应)。通过调用适当的执行引擎，我们将提供以逻辑时间(通常的模拟时钟)和实时(受限于挂钟时间)执行模型的能力。

Real-Time Embedded systems are built as sets of components interacting with their surrounding environment (transport, automated buildings, etc.). Their correctness is critical, as well as their timeliness: the moment when system tasks are executed can lead to catastrophic consequences (for goods or lives). We propose advance researching the use of Modeling and Simulation in real-time embedded environments. We propose use a formal modeling and simulation approach as formalized methods can provide good results because the software artifacts used for real-time modeling and simulation are of higher quality. In particular, we focus on the application of the DEVS formalism, a formal modeling and simulation framework based on generic dynamic systems concepts. We propose using DEVS for Real-time systems, using DEVS theory. Although the state-of-the-art in DEVS modeling has advanced steadily, none of the existing DEVS research results applied in real-time contexts consider a variety of complex problems in real-time applications: fault tolerance, sensor replication transient system overloading, and others. We will advance in new techniques to solve this problem, based on the use of imprecise computation and scheduling methods, presenting new methods for controlled degradation under transient overloads (by dividing the tasks in a mandatory part to provide basic results, and an optional part to improve the results obtained; also multiple versions of the same task can be provided, using a different version depending on the present load). The goal of our research program is to combine various theoretical frameworks, providing a technique for imprecise real-time computing based on DEVS. From the methodological point of view, we are interested in developing a theory of dynamic RT DEVS using imprecise computing. The goal is to provide security, reduction in the development times, and the advantages of a formal framework to develop simulations. The results of the simulated system can be applied in the development in the real-time system. From the practical point of view, we expect to provide a set of tools that can be applied to develop real-time software and simulations with hardware-in-the-loop (requiring real-time response). By invoking appropriate execution engines, we will provide the ability to execute models in both logical time (the usual simulation clock) and real time (constrained to wall-clock time).