Methodologies for real-time discrete-event simulation

点击翻译按钮获取中文摘要

Real-time systems (RTS) are very advanced computer system applications with hardware and software components interacting in a tight fashion. The manual development of this kind of systems is time consuming, error prone, and expensive. Different formal methods have been proposed for RTS development, but they have shown to be difficult to apply as the complexity of the system scales up. Modeling and Simulation (M&S) techniques and tools have been proposed to address this issue and have been proven to be useful to verify and validate RTS, including the environment they interact with. We want to investigate systematic methods and automated tools to develop RTS using a model-based methodology. We are interested in defining new theories, methods and techniques for building simulation models that can later be applied in real-time environments, and in systems integrating complex physical systems with RTS, combining formal modelling techniques and simulation. Although several efforts exist in this area, many questions remain open. The main issue is that none of the proposed solutions has considered problems of transient overloading, checking of timing constraints, and fault tolerance. This requires defining new theories, algorithms and tools to guarantee responses to inputs within specified time constraints, supporting faults and providing graceful degradation. We will also analyze how to define real-time models, and we will provide automatic verification for the executing models, including methods for continuous and hybrid systems. We will build a set of tools that can be applied to develop real-time software, and simulations with hardware in-the-loop, which will be used as a proof-of-concept of our theoretical framework. The research has the potential to influence a variety of applications fields (communication, emergency planning, traffic control, biomedical, etc).