Model-based Test Generation for Embedded Software with Simulink/Stateflow or Hybrid Automata Models

提案编号：0811541标题：使用 Simulink/Stateflow 或混合自动机模型为嵌入式软件生成基于模型的测试 PI：Ratnesh Kumar 和 Changyan Zhou，电气系。 ＆比较。爱荷华州立大学艾姆斯分校的工程师提出了一种基于 IAA 模型的测试生成方法，适用于反应性（与其环境交互）、具有混合行为（与切换逻辑混合的连续动态）并且可以使用 Simulink/Stateflow 或混合自动机模型进行建模的嵌入式软件系统。 Simulink/Stateflow是一种用于嵌入式软件开发的图形化建模、分析和仿真工具，广泛应用于汽车、飞机、医疗器械等行业。这项研究的智力价值包括：(i) 将 Simulink/Stateflow 模型转化为混合自动机模型（尽可能），(ii) 基于离散行为保留抽象和系统识别技术的混合系统测试生成，以及 (iii) 开发与 Matlab Simulink/Stateflow 环境集成的支持软件。拟议的研究可能会对 Simulink/Stateflow 通常用于嵌入式软件开发的应用领域产生影响。作为该项目的一部分开发的一些技术可能会应用于其他领域，例如故障识别的验证或监控。该项目将支持两名新研究人员，他们将接受基于模型的嵌入式软件测试方法的培训。研究成果，包括软件工具，将通过 PI 的主页和期刊/会议出版物提供。 PI 致力于为该项目招募少数族裔学生。

Proposal No.: 0811541Title: Model-based Test Generation for Embedded Software with Simulink/Stateflow or Hybrid Automata ModelsPI: Ratnesh Kumar and Changyan Zhou, Dept. of Elec. & Comp. Eng., Iowa State Univ., Ames, IAA model-based test generation approach for embedded software systems that are reactive (interact with their environment), possess hybrid behaviors (continuous dynamics mixed with switching logics), and can be modeled using Simulink/Stateflow or hybrid automata models, is proposed. Simulink/Stateflow is a graphical modeling, analysis, and simulation tool for embedded software development, and is widely used in industries such as automobile, aircraft, and medical devices. The intellectual merits of this research include: (i) Translation of Simulink/Stateflow models into hybrid automata models (to the extent possible), (ii) Test generation for hybrid systems based on discrete-behaviors preserving abstractions and system identification techniques, and (iii) Development of supporting software to be integrated with the Matlab Simulink/Stateflow environment. The proposed research is likely to have an impact in application domains where Simulink/Stateflow is routinely used for embedded software development. Some of the techniques developed as part of the project may find application in other areas such as verification or monitoring for fault identification. The project will support two new researchers who will be trained in model-based approaches to embedded software testing. The research outcomes, including software tools, will be made available through PIs' homepages and journal/conference publications. The PIs are committed to recruiting minority students for the project.