Robust and autonomous machine learning on board a miniature satellite

微型卫星上的鲁棒自主机器学习

• 批准号: NI220100072
• 负责人: Prof James Bailey
• 金额: $ 39.34万
• 依托单位: The University of Melbourne
• 依托单位国家: 澳大利亚
• 项目类别: National Intelligence and Security Discovery Research Grants
• 财政年份: 2022
• 资助国家: 澳大利亚
• 起止时间: 2022-07-01 至 2025-06-30
• 项目状态: 未结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/NI220100072
• 关键词: Robust; autonomous; machine; learning; board

The advent of miniature satellites (CubeSats) orbiting in space is a new and exciting frontier. There is an emerging opportunity for CubeSats to use adaptable machine learning models to perform on board analysis and processing of data. This proposal aims to research, develop and deliver state of the art approaches that tackle autonomy and robustness challenges for machine learning within a CubeSat. This will enable machine learning models to operate in the hostile and low resource environment of space. A distinguishing benefit of this project is our plan to deploy and evaluate new machine learning approaches on board a soon to be launched CubeSat, demonstrating Australia’s leadership at an international level.

在太空轨道运行的微型卫星（CubeSats）的出现是一个令人兴奋的新领域。CubeSats有一个新的机会，可以使用适应性强的机器学习模型来进行数据的机载分析和处理。 该提案旨在研究，开发和提供最先进的方法，以解决CubeSat内机器学习的自主性和鲁棒性挑战。 这将使机器学习模型能够在太空的恶劣和低资源环境中运行。 该项目的一个显著好处是，我们计划在即将推出的CubeSat上部署和评估新的机器学习方法，展示澳大利亚在国际层面的领导地位。