I-Corps: Advancing Space Computing

I-Corps：推进空间计算

• 批准号: 2347117
• 负责人: Brock LaMeres
• 金额: $ 5万
• 依托单位: Montana State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-11-15 至 2024-10-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2347117&HistoricalAwards=false
• 关键词: Corps; Advancing; Space; Computing

The broader impact/commercial potential of this I-Corps project is the development of a radiation-tolerant computer for space and other applications. Radiation tolerance of electronic components is their ability to shield themselves from the radiation in their working environments. A single charged particle can knock thousands of electrons loose, causing electronic noise and signal spikes. In the case of digital circuits, this can cause results that are inaccurate or unintelligible. The goal for the proposed technology is to achieve the necessary levels of space computing technology reliability in the presence of radiation at a fraction of the cost of existing systems. The proposed technology is suitable for command and data handling tasks for small spacecraft that require lower performance but desire fault-tolerance. Modifications may be made to increase its performance by adding hardware accelerators. This higher performance version may be suitable for military and defense applications. The inclusion of malware mitigation may make the proposed technology applicable to critical infrastructure applications such as in the nation’s electrical grid, water treatment plants, and hospitals. Providing more reliable computers in these systems may lead to a more resilient infrastructure that will not fail when under stress.This I-Corps project is based on the development of a computer that is able to continue foreground operation in the presence of radiation-induced faults and repair faulted circuity in the background. The proposed technology takes advantage of reconfigurable commercial off-the-shelf (COTS) parts instead of “hardening” the underlying computer materials to prevent faults from occurring. The proposed approach expects faults to occur and provides a recovery procedure to mitigate their impact. By avoiding the expensive hardening process and using COTS parts, the cost of the proposed technology is reduced over existing solutions. The architecture used to recover from faults simultaneously provides the ability to embed advanced algorithm accelerators in the computer to increase its computational performance with greater power efficiency than existing solutions. This approach has been tested on numerous space demonstrations including small satellites, the International Space Station, and in 2024 it will travel to the lunar surface for its harshest test yet. The advancement in computation provided by this technology may enable future space missions with greater science return and deeper exploration of the galaxy by increasing spacecraft autonomy and real-time science data processing. In addition, an additional layer of fault intrusion defense obfuscates the computing hardware so that malware cyberattacks may be detected and defeated. This extends the use of this technology beyond space and into terrestrial applications such as the nation’s critical infrastructure that must withstand both environmentally induced faults and human initiated cyberattacks.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

I-Corps项目的更广泛影响/商业潜力是开发一种用于空间和其他应用的耐辐射计算机。 电子元器件的辐射耐受性是指其在工作环境中屏蔽辐射的能力。单个带电粒子可以使数千个电子松动，导致电子噪声和信号尖峰。在数字电路的情况下，这可能导致结果不准确或难以理解。拟议技术的目标是在存在辐射的情况下以现有系统的一小部分成本实现必要水平的空间计算技术可靠性。 所提出的技术适用于小型航天器的命令和数据处理任务，要求较低的性能，但希望容错。 可以进行修改以通过添加硬件加速器来提高其性能。 这种更高性能的版本可能适用于军事和国防应用。 包含恶意软件缓解可以使所提出的技术适用于关键基础设施应用，例如国家电网、水处理厂和医院。在这些系统中提供更可靠的计算机可能会导致更有弹性的基础设施，在压力下不会出现故障，这个I-Corps项目的基础是开发一台计算机，该计算机能够在出现辐射引起的故障时继续前台操作，并在后台修复故障电路。 所提出的技术利用可重新配置的商用现货（COTS）部件，而不是“硬化”底层计算机材料，以防止故障发生。 所提出的方法预计故障发生，并提供了一个恢复程序，以减轻其影响。 通过避免昂贵的硬化过程和使用COTS部件，所提出的技术的成本比现有的解决方案降低。 用于从故障中恢复的架构同时提供了在计算机中嵌入高级算法加速器的能力，以提高其计算性能，并具有比现有解决方案更高的功率效率。这种方法已经在包括小型卫星和国际空间站在内的众多太空演示中进行了测试，并将于2024年前往月球表面进行迄今为止最严格的测试。 这项技术提供的计算进步可能使未来的空间任务能够通过提高航天器的自主性和实时科学数据处理来获得更大的科学回报和更深入的星系探索。 此外，故障入侵防御的附加层使计算硬件模糊，使得可以检测和击败恶意软件网络攻击。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。