Data-driven behavioural modelling for military applications

This article investigates the possibilities for creating behavioural models of military decision making in a data-driven manner. As not much data from actual operations is available, and cannot easily be created context, most approaches use simulators to learn behaviour. A simulator however always available or difficult create. study focusses on creation that was collected during field exercise. general limited, noisy erroneous, this makes realistic challenging. Besides using traditional approach hand-crafting model based data, we investigate emerging research area imitation learning. One its techniques, reward engineering, applied behaviour soldiers an urban warfare operation. Basic, but realistic, soldier learned, which lays groundwork more elaborate future.