SBIR Phase I: Debugging Smart Cyberphysical Systems

这个小企业创新研究（SBIR）第一阶段项目的更广泛的影响/商业潜力基于两个趋势：（a）软件正在迅速成为工程产品（如汽车、医疗设备和无人机）中最复杂和最昂贵的部分;（B）现有的测试和模拟技术不足以在早期设计中发现软件错误。这导致昂贵的召回，有时甚至是致命的故障，越来越多的认证机构正在纳入严格的测试和验证标准。该项目的成功完成将产生一个可扩展的原型工具，这将有助于向我们的潜在客户（即工程系统设计师）证明，在不影响质量的情况下，可以从根本上降低软件开发成本。使用我们的算法和工具，可以有效地早期发现错误，并提供满足设计要求的证书。小企业创新研究（SBIR）第一阶段项目旨在开发可扩展分析设计正确性的商业技术。该项目中内置的软件工具将通过快速数值模拟与设计模型的自动分析相结合来自动检查系统设计的安全性，以涵盖少量痕迹中的大量可能行为。我们的技术在汽车、航空电子和医疗设备行业的几个行业规模的挑战问题上取得了明显的成功。在这个项目中，我们将开发一个抛光的原型，我们的软件工具的功能，将提高设计师的经验。它将为调试复杂的系统设计提供自动化支持。这些功能将使该工具的学习曲线变平，并使该方法适用于现有的行业标准设计环境，如Simulink/Stateflow。我们将为该工具开发用例，以展示针对具有挑战性问题的整个设计工作流程。原型和用例将成为创建最小可行产品和beta测试的基础。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is based on two trends: (a) software is quickly becoming the most complex and expensive part of engineering products like cars, medical devices, and drones and (b) existing testing and simulation techniques are inadequate for finding software bugs in early design. This leads to expensive recalls and sometimes fatal failures, and increasingly, certification bodies are incorporating stringent testing and verification standards. Successful completion of this project will result in a scalable prototype tool that will help demonstrate to our potential customers, namely the engineering system designer, that it is possible to radically cut down software development cost without compromising on quality. This will be possible using our algorithms and tools that enable effective early discovery of bugs as well as coverage guarantees that give certificates for meeting design requirements. This Small Business Innovation Research (SBIR) Phase I project aims to develop commercial technology that can scalably analyze the correctness of designs. The software tool built in this project will automatically check the safety of system designs by combining fast numerical simulations, with automatic analysis of design models to cover a large set of possible behaviors from a few traces. Our techniques have had demonstrable success on several industry-scale challenge problems in the automotive, avionics, and medical devices industries. In this project, we will develop a polished prototype of our software tool with features that will enhance the designer's experience. It will provide automated support for debugging complex system designs. These features will flatten the learning curve for the tool and make the approach applicable to existing industry standard design environments like Simulink/Stateflow. We will develop use-cases for the tool that will demonstrate the entire design workflow on challenging problems. The prototype and the use cases will become the basis for the creation of a minimum viable product and for beta testing.