Collaborative Research: SoD-TEAM: A Feedback-Based Architecture for Highly Reliable Embedded Software

合作研究：SoD-TEAM：基于反馈的高度可靠嵌入式软件架构

• 批准号: 0650049
• 负责人: Ella Atkins
• 金额: $ 5万
• 依托单位: Regents of the University of Michigan - Ann Arbor
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2006
• 资助国家: 美国
• 起止时间: 2006-09-01 至 2008-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0650049&HistoricalAwards=false
• 关键词: Collaborative; Research; SoD; TEAM; Feedback

Title: "Collaborative Research: SoD-TEAM: A Feedback-Based Architecture for Highly Reliable Embedded Software"The focus of this project is on developing a theory and architectural framework for incorporating feedback control as a main principle of software engineering. This proposal develops scientific foundations, tools, and architectural design principles to improve the reliability and reduce the development cost of mission-critical software. Two common approaches to ensure reliable system behavior are: 1) Scientific foundations, tools, and architectural principles for ensuring elimination of errors, and 2) foundations, tools, and architectural principles for ensuring tolerance to such errors. While the first solution is sufficient, building completely error-free large-scale systems has been an elusive exercise as evidenced by experiences with today's engineering artifacts (e.g., recalls on vehicles, cascading power blackouts, etc). Consequently, a new approach to robust software design, development and certification is needed that addresses the increasing cost and safety implications of the current practices. Feedback control has proven to be a very successful tool for ensuring correct behavior of complex, poorly modeled systems in the presence of uncertainty. This project applies principles of feedback control and stability envelopes to the engineering of software systems.Program Manager: Anita J. La SalleDate:

标题：“协作研究：SOD-TEAM：基于反馈的架构，用于高度可靠的嵌入式软件”该项目的重点是开发一个理论和体系结构框架，用于将反馈控制纳入软件工程的主要原则。该建议开发了科学基础，工具和建筑设计原则，以提高可靠性并降低关键任务软件的发展成本。确保可靠系统行为的两种常见方法是：1）确保消除错误的科学基础，工具和建筑原则，以及2）基础，工具和建筑原则，以确保对此类错误的容忍。虽然第一个解决方案就足够了，但构建完全无错误的大规模系统一直是一种难以捉摸的练习，这可以通过当今的工程文物（例如，召回车辆，级联电源停电等）证明。因此，需要一种新的用于强大的软件设计，开发和认证的方法，以解决当前实践的成本和安全性的不断增加。事实证明，反馈控制是确保在不确定性存在下确保复杂，建模较差的系统正确行为的非常成功的工具。该项目将反馈控制和稳定信封的原理应用于软件系统的工程。