Autonomous Navigation of UAVs in Complex Battlefield Environments Using Ant Colony Optimization

The autonomous navigation of UAVs in complex battlefield environments represents a cutting-edge research topic. Our approach implements Ant Colony Optimization (ACO) algorithm through MATLAB simulation, modeling path planning with pheromone-based probability transitions between nodes. The UAV operates within a 20×20 km area with initial coordinates [1,2] km and target destination [19,18] km, excluding takeoff/landing phase constraints. The digital terrain map (provided in attachments) incorporates radar threat modeling and simplifies topography into three classifications: highland, lowland, and transition zones. For single-UAV single-target trajectory planning, we developed a 2D planning model that prioritizes threat avoidance and range optimization while excluding terrain and maneuverability constraints. The MATLAB implementation features key functions including: 1) Pheromone matrix initialization for path reinforcement learning, 2) Probability-based node selection using roulette wheel selection, and 3) Dynamic threat cost calculation integrating radar detection probabilities. This baseline model can be extended to incorporate additional real-world factors such as weather conditions and 3D terrain features through modular code expansion in the objective function calculation module.