Modern and future battlefields are evolving into environments characterized by increasing uncertainty and complexity, making it difficult for commanders to make swift and accurate command decisions. For commanders of combat units below the battalion l...
Modern and future battlefields are evolving into environments characterized by increasing uncertainty and complexity, making it difficult for commanders to make swift and accurate command decisions. For commanders of combat units below the battalion level, where staff support is limited, an intelligent command decision support system capable of providing accurate situational awareness and effective countermeasures is essential. This paper proposes an intelligent command decision support system that integrates and analyzes diverse multimodal battlefield data collected in real time, enabling accurate judgment and optimal response plans. To achieve this, a battlefield ontology-based knowledge base and knowledge graph were built, and an AI-based inference model that combines the ComplEx embedding model with a relational graph convolutional neural network (R-GCNN) was implemented. The proposed model learns complex relationships within the battlefield knowledge graph, and infers new knowledge, enabling high-level situational awareness and decision-making support even in information-limited environments. The results of this study are expected to contribute to a paradigm shift in command and control by overcoming the limitations of existing command decision-making processes, which rely on an individual commander's experience and intuition. Furthermore, the model will likely evolve into a core component of future intelligent command and control systems.