Dynamic reconnaissance operations with UAV swarms: adapting to environmental changes.

This study introduces a novel framework for dynamic reconnaissance operations using Unmanned Aerial Vehicle (UAV) swarms, designed to adapt in real time to changes in mission parameters and UAV availability. Unlike traditional models that assume static operational conditions, our approach distinguishes between two key categories of change: Type I, related to modifications in the UAV swarm (e.g., vehicle loss or deployment), and Type II, concerning adjustments in mission configuration or the area of responsibility. These are jointly addressed within a unified optimization framework based on Ant Colony Optimization (ACO), allowing efficient trajectory planning and rapid replanning during mission execution. As part of the framework, we propose a Pheromone Matrix Initialization (PMI) technique to accelerate convergence in Type I scenarios by reusing heuristic information from prior optimizations. The effectiveness of the overall framework is validated through six realistic scenarios, demonstrating its ability to maintain mission continuity with minimal delay and to respond efficiently to complex and sequential changes. Comparative analysis shows consistent superior performance over classical and state-of-the-art methods, with reductions in optimization time and mission completion time. This work delivers a practical, scalable solution for mission planning in uncertain and time-sensitive UAV operations.