Abstract

Military planning operations require navigating constantly changing environments. To support decision-makers, innovative concepts are essential for automatically generating effective solutions tailored to specific logistics operations. These tools aim to accelerate planning procedures, minimize risks, and decrease operating costs. This paper introduces a simulation-based optimization framework designed to enhance the mobility of military vehicles through terrain-aware navigation. The paper specifically delves into a key component of the framework: terrain identification. This challenge is addressed using unsupervised methods, ensuring applicability even in unfamiliar operational settings. The experimental findings demonstrate promising results in identifying terrain characteristics, particularly in discerning surface waviness, slant, and curvature.