DQN-ALNS for UAV Reconnaissance Routing Problem with Target Priority Consideration

<strong>&nbsp;</strong>In recent years, UAVs have been extensively employed for reconnaissance missions. Our focus is &nbsp;prioritizing reconnaissance of high-priority targets while minimizing the flight duration when UAV power is &nbsp;constrained. We introduce a framework called DQN-ALNS, which integrates Deep Q Network (DQN) into the &nbsp;Adaptive Large Neighborhood Search (ALNS) metaheuristic algorithm to optimize the process through the &nbsp;current solution’s search state. Specifically, the agent is utilized to select the destroy-repair operators to update &nbsp;a new solution, thereby iteratively optimizing the UAV reconnaissance routes. Experimental results reveal that &nbsp;DQN-ALNS achieves superior solutions and faster convergence than other comparison algorithms. The &nbsp;algorithm leverages the exploratory potential of the current solution and demonstrates robust stability. The final &nbsp;sensitivity analysis showcases that reconnaissance missions with high priority are better accomplished when &nbsp;the UAV power is moderate and the target priority is concentrated at smaller values.&nbsp;&nbsp;