<![CDATA[The movement of combat vehicles in modern battlefields faces complex challenges in the form of uncertain terrain, dynamic enemy threats, and limited real-time information, making conventional methods such as Dijkstra or A* less capable of optimising routes adaptively. This research aims to develop an Adaptive Ant Colony Optimization (ACO) algorithm model integrated with a dynamic risk map to determine safe, fast, and efficient routes for combat vehicles. The methodology employed includes designing an adaptive ACO with risk-based pheromone update mechanisms, modeling dynamic risk maps using Gaussian probability functions and Markov models, and conducting graph-based battlefield simulations to evaluate algorithm performance. Evaluation was conducted by comparing the adaptive ACO with baseline algorithms (Dijkstra, A*, and Particle Swarm Optimization) using metrics such as Safety Index (SI), Time Efficiency (TE), Adaptability, and Computational Cost (CC). The results show that the adaptive ACO consistently produces paths with the highest SI values, competitive time efficiency, and better real-time adaptability compared to the baseline, while path visualization demonstrates the algorithm's ability to dynamically avoid high-risk areas. These findings indicate that integrating adaptive ACO with dynamic risk maps provides safer and more flexible navigation strategies, with significant potential for application in autonomous combat vehicles, UAV systems, and military operations based on intelligent simulation. This research contributes to the development of adaptive path optimization algorithms in dynamic battlefields, bridges the literature gap related to risk-based navigation, and provides a framework that can serve as the foundation for developing military decision support systems based on artificial intelligence. ]]>
