SBIR Phase I: Multi-Sensor 12U Spacecraft with On-Board Data Fusion and Processing

这一小企业创新研究第一阶段项目的更广泛影响将是使新一代小卫星能够协助许多基本功能，从环境管理和执法到自然灾害评估和城市研究，提高林业和农业产量，石油和天然气管道检查，排放和海洋污染监测，土地使用调查和规划，以及许多其他商业、工业、环境和人道主义用例。 该技术将使人们能够进入价值100亿美元的新空间应用市场。 拟议的项目将开发一种通过一体化机载数据融合和处理进行小型卫星星座运行的新方法。 该技术是首个与基于神经网络技术的图像处理和目标检测技术紧密结合的高有效载荷总线。 这是一种在卫星本身上创建“边缘智能”的深层方法。该项目将：1）使用建筑物检测作为概念性示例来演示融合来自多个源的数据的概念验证深度学习系统（在这种情况下，可见光RGB和短波红外图像）; 2）评估数据融合和处理算法在小尺寸上的性能和兼容性，重量和功率（SWaP）处理器，专为太空而定制-该奖项反映了NSF的法定使命，并被认为值得通过使用基金会的智力价值和更广泛的影响进行评估来支持审查标准。

The broader impact of this Small Business Innovation Research (SBIR) Phase I project will be to enable a new generation of small satellites to aid many essential functions ranging from environmental management and law enforcement to natural disaster evaluation and urban studies, improving forestry and agricultural yields, oil and gas pipeline inspection, emissions and marine pollution monitoring, land-use surveys and planning, and a host of other commercial, industrial, environmental and humanitarian use-cases. The technology will enable access to a $10B addressable market of new space applications. The proposed project will develop a new approach to small satellite constellation operations through integrated on-board data fusion and processing. The technology is a first-of-its-kind, high payload volume bus closely coupled with image processing and object detection techniques based on neural net technologies. This is a deeply layered approach that create “edge-intelligence” on the satellite itself. This project will: 1) use building detection as a notional example to demonstrate a proof-of-concept deep learning system fusing data from multiple sources (in this case, visible RGB and short wave infra-red imagery); 2) assess the performance and compatibility of the data fusion and processing algorithms on a low size, weight and power (SWaP) processor customized for space-based applications.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.