CSR-EHS: Software Composition for Embedded Systems using Graph Transformations

基于模型的软件开发方法有望显著提高生产力。然而，目前模型驱动架构（MDA）的工业努力普遍缺乏严格的基础，其嵌入式系统开发的全部潜力尚未得到证明。本课题的探索性研究主要围绕以下几个方面展开：（1）. 该项目为基于特定领域建模语言（DSML）的具体MDA方法开发了正式的基础。 探索性研究的一个核心组成部分是调查图转换系统是否可以为实际的模型转换和软件生成提供一个合适的技术框架。（二）、 该项目通过为嵌入式应用程序构建专门的DSML-s，并演示如何从相同的代码库为不同的平台生成软件，来应用分布式实时嵌入式系统领域的研究成果。该项目旨在通过探索：嵌入式应用程序的特定领域模型，形式化模型转换和模型的平台相关软件合成来提高构建复杂嵌入式软件系统的知识。这项研究的更广泛的社会影响在于构建和维护安全可靠的复杂实时嵌入式系统应用程序的严格方法的重要性。

Model-based approaches to software development promise significant increase in productivity. However, current industrial efforts in Model Driven Architectures (MDA) generally lack rigorous foundations, and their full potential for embedded system development is yet to be demonstrated. The exploratory research in this project addresses these issues in the following research thrusts:(1). The project develops formal foundations for a concrete MDA approach based on Domain-Specific Modeling Languages (DSMLs). A central component of the exploratory research investigates whether graph transformation systems can provide a suitable technology framework for practical model transformation and software generation.(2). The project applies the results of research in the domain of distributed, real-time embedded systems by building specialized DSML-s for embedded applications, and demonstrating how software can be generated for different platforms from the same code base. The project seeks to advance knowledge about constructing complex embedded software systems by exploring: domain-specific models for embedded applications, formal model transformation, and platform-dependent software synthesis from models. The broader, societal impact of this research lies in the importance of rigorous methods for building and maintaining complex real-time embedded system applications that are safe and reliable.