Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB 
Intelligent Information Management for Coalition Shared Data (CSD)
In joint multinational operations, it is crucial to provide relevant information on demand and in a timely manner. In the Joint ISR domain, reconnaissance results are shared as Joint ISR products. Coalition Shared Data (CSD) servers compliant with NATO STANAG 4559 enable the distributed storage and dissemination of these products. Within a network of CSD servers, Joint ISR products can be discovered via their metadata and subsequently retrieved for further use.
Multimodal Semantic Content Search
To improve access to decision-relevant information in Joint ISR products and to enable further processing, we are developing AI systems based on LLMs and Retrieval-Augmented Generation (RAG). Our ARGoS approach enhances the existing metadata search in CSD servers with a unified multimodal search over product content. This enables users to search directly for relevant information within, for example, reconnaissance reports and imagery.
Need-based preparation to support analysis and situation reports
ARGoS not only enables semantic search across Joint ISR products but also supports the extraction of specific content for AI-assisted further processing and refinement. In combination with LLMs, it can generate tailored summaries and analyses, for example to contribute to situational awareness.
Independence through locally operated AI models
ARGoS consistently relies on local data storage and the use of locally operated AI models to maximize data sovereignty and address security requirements.
