Offline Language-Based Support System for Tactical Decision-Making in Air Combat Operations

Abstract

An offline language-based support tool is presented to assist pilot training and tactical decision-making in air combat operations. The system applies retrieval-augmented generation (RAG) to answer natural language questions using content extracted from doctrinal and tactical manuals. Built entirely with open-source components, it performs text preprocessing, embedding generation, and semantic search locally, ensuring full offline functionality. Current capabilities are demonstrated through use cases involving air combat manuals, and potential applications are discussed in onboard mission systems, flight simulators, and training environments. The source code is modular and intended for public release to support future extensions and integration.

@inproceedings{dantas2025offline,
  author = {Dantas, Joao P. A.},
  title = {{Offline Language-Based Support System for Tactical Decision-Making in Air Combat Operations}},
  booktitle = {Conference on Neural Information Processing Systems (NeurIPS), LatinX in AI (LXAI) Workshop},
  year = {2025}
}