Pentagon looks to AI to identify space and missile threats

Missile defense in wartime involves many complications. Human operators must deal with high-speed objects, short timelines and sensor data that isn’t always clear.

The Pentagon now wants artificial intelligence that can cut through the confusion.

It’s seeking “AI-based software tools to turn fragmented sensor data into a continuously updated understanding of space and missile threats fast enough to drive operational decisions,” according to the Defense Innovation Unit solicitation , which is due Sept. 24. “Existing tools struggle to distinguish closely spaced objects, track emerging threats, and keep threat models current.”

The project, known as the “Space Threat Intelligence Synthesis Engine,” will leverage AI’s greatest strength: the ability to quickly analyze reams of data. The software will fuse “massive streams of multi-source data, including live video, satellite imagery, radar and sensor feeds, geospatial data, and classified intelligence reports.”

“By automating advanced cross-data correlation, the architecture aims to uncover hidden patterns, complex operational relationships, and predictive threat behaviors far beyond the capabilities of human analysis alone.”

Ultimately, the system should “generate actionable, confidence-scored alerts supporting analyst-in-the-loop, human-on-the-loop, or automated workflows.”

The AI will help human operators — and other machines.

It will provide “intuitive, human-readable visualizations for frontline operators and low-latency, machine-to-machine APIs to drive automated command-and-control (C2) workflows.”

DIU is aiming for an open-source system with a latency delay of no more than five seconds — and preferably no more than two seconds — between the arrival of data and the results being displayed. Throughput should be 20 to 30 megabytes per minute, with bursts of up to five gigabytes.

Contractors should demonstrate that their products show “significant speed and accuracy improvements in event detection, characterization, and attribution compared to existing baseline systems.”

As skies fill with missiles and drones in conflicts like the ones with Ukraine and Iran, AI is increasingly being seen as an indispensable partner to overworked human operators. In particular, with a perpetual shortage of expensive interceptors , such as the Patriot air-defense system, operators must quickly decide when and where to allocate them.

Defense contractors and AI firms are teaming up to incorporate AI into missile defense.

For example, Northrop Grumman announced in July that it was partnering with AI firm Camgian to incorporate artificial intelligence into integrated air and missile defense systems. BAE and Scale AI have also teamed up for AI-enhanced missile defense.

Michael Peck is a correspondent for Defense News and a columnist for the Center for European Policy Analysis. He holds an M.A. in political science from Rutgers University. Find him at theuncommondefense.com. His email is mikedefense1@gmail.com.

战时导弹防御涉及诸多复杂因素。操作人员必须应对高速目标、紧迫的时间限制以及并非总是清晰的传感器数据。

五角大楼现在需要一种能够消除混乱局面的人工智能。

根据国防创新部门的招标书（截止日期为9月24日），该部门正在寻求“基于人工智能的软件工具，以将零散的传感器数据转化为对太空和导弹威胁的持续更新的理解，从而能够快速地驱动作战决策”。“现有工具难以区分近距离物体、跟踪新出现的威胁以及保持威胁模型的更新。”

该项目名为“太空威胁情报合成引擎”，将充分利用人工智能的最大优势：快速分析海量数据的能力。该软件将融合“来自多个来源的大量数据流，包括实时视频、卫星图像、雷达和传感器数据、地理空间数据以及机密情报报告”。

“通过自动化高级跨数据关联，该架构旨在发现隐藏的模式、复杂的操作关系以及预测威胁行为，这些远远超出了仅靠人类分析的能力。”

最终，该系统应“生成可操作的、置信度评分的警报，以支持分析师参与、人工参与或自动化工作流程”。

人工智能将帮助人类操作员以及其他机器。

它将为一线操作员提供“直观、易于理解的可视化效果，以及低延迟的机器对机器 API，以驱动自动化指挥与控制 (C2) 工作流程”。

DIU的目标是开发一个开源系统，数据到达和结果显示之间的延迟不超过五秒，最好不超过两秒。吞吐量应为每分钟20至30兆字节，峰值可达五千兆字节。

承包商应证明，与现有基线系统相比，他们的产品在事件检测、特征描述和归因方面具有“显著的速度和准确性提升”。

在乌克兰和伊朗等冲突中，导弹和无人机充斥着天空，人工智能正日益被视为过度劳累的人工操作员不可或缺的伙伴。尤其是在像“爱国者”防空系统这样昂贵的拦截器长期短缺的情况下，操作员必须迅速决定何时何地部署它们。

国防承包商和人工智能公司正在携手合作，将人工智能技术应用于导弹防御系统。

例如，诺斯罗普·格鲁曼公司在7月份宣布与人工智能公司Camgian合作，将人工智能融入一体化防空和导弹防御系统。英国航空航天系统公司（BAE Systems）和Scale AI公司也合作开发人工智能增强型导弹防御系统。

迈克尔·佩克是《防务新闻》的通讯员，也是欧洲政策分析中心的专栏作家。他拥有罗格斯大学政治学硕士学位。您可以在theuncommondefense.com找到他。他的邮箱是mikedefense1@gmail.com。