字号·· | 护眼
商业内幕

乌克兰希望用AI预测俄罗斯无人机将何时何地袭击前线Ukraine wants AI to predict where and when Russian drones will hit the front lines

点「原文对照」整页切到原文,或双击某段只看那段的原文。

在扎波罗热地区,乌克兰国家警察斯特列尔茨基特种部队营的一名士兵在一门乌克兰2S22“博赫达纳”155毫米自行榴弹炮旁看着智能手机。A soldier from the Striletskyi special forces police battalion of the National Police in the Zaporizhzhia region looks at a smartphone near a Ukrainian 2S22 Bohdana 155 mm self-propelled howitzer.

乌克兰国防部周二表示,其部队即将开始使用北约共同开发的人工智能模型来预测俄罗斯无人机袭击的落点和时间。该模型被命名为“模型13号”,它通过分析过去的作战数据、天气条件和战场情况,以“高概率”确定俄罗斯军队将如何在前线使用无人机,国防部在一份声明中表示。该模型可以预测袭击的时间、地点、强度以及所使用的无人机类型,国防部称。

Ukraine's defense ministry said on Tuesday that its forces are about to start using a jointly NATO-developed artificial intelligence model to predict where and when Russia's drone strikes will land. Dubbed "Model No. 13," the tool analyzes past combat data, weather conditions, and battlefield conditions to determine, with "high probability," how Russian troops will use drones on the front lines, the ministry said in a statement. The model can forecast the strike's time, location, intensity, and type of drone used, the ministry said.

声明写道:“该解决方案已经使用真实世界数据集进行了验证。它已被改进,并准备好融入乌克兰武装部队的态势感知系统。”国防部表示,模型13号是由其武装部队和北约盟军司令部转型司令部共同开发的,并得到了联合分析、培训和教育中心的支持,该中心是北约在2024年底成立的组织,旨在根据战争中的经验教训推动创新。

"The solution has already been validated using real-world datasets. It has been refined and is ready for integration into the Armed Forces of Ukraine's situational awareness systems," its statement read. The ministry said Model No. 13 was developed by its armed forces and NATO's Allied Command Transformation, with support from the Joint Analysis, Training and Education Center, an organization the alliance created in late 2024 to help it innovate based on lessons learned in the war.

ACT和JATEC尚未对商业内幕在正常工作时间之外发送的置评请求作出即时回应。乌克兰国防部表示,另一个用于规划炮兵部署的人工智能模型——32号模型——也正在共同研发中。然而,乌克兰军队仍在测试该模型的原型,国防部补充道。此次公告是在乌克兰官员2月份表示其部队自去年以来一直在使用人工智能和机器学习模型处理战争中的作战数据以进行“威胁预测”之后发布的。

ACT and JATEC did not immediately respond to requests for comment sent outside regular business hours by Business Insider. Another AI model that plans artillery deployments, No. 32, is also being jointly created, Ukraine's defense ministry said. However, Ukrainian troops are still testing a prototype of the model, it added. The announcement comes after Ukrainian officials said in February that their forces had been working with AI and machine learning models since last year to process combat data from the war for "threat forecasting."

模型13号很可能将在乌克兰的DELTA战场管理系统上运行,这是一款数字应用程序,能够整合来自无人机、摄像头和战场报告的情报,为指挥官提供前线概览,他们可以通过平板电脑或手机查看。

Model No. 13 is likely to run on Ukraine's DELTA battlefield management system, a digital app that combines intelligence from drones, cameras, and battlefield reports to give commanders an overview of the frontline from their tablets or phones.

该系统已经使用了一个名为Vezha的AI模块,乌克兰称该模块能够通过分析卫星影像、无人机摄像头画面和无线电信号,在几秒钟内自动检测并标记俄罗斯装备。乌克兰国防部表示,DELTA系统在11月份的用户数量已超过20万。

The system already uses an AI module called Vezha that Ukraine says can automatically detect and flag Russian equipment in seconds by analyzing satellite imagery, drone camera feeds, and radio signals. DELTA reached over 200,000 users in November, Ukraine's defense ministry has said.