The Application of Hybrid GA with Rule-Based Population Initialization in UAV Refueling Schedule

Unmanned Aerial Vehicle (UAV) refueling scheduling is critical for time-sensitive missions, especially in military operations where traditional return-to-base refueling is impractical. This paper proposes a hybrid genetic algorithm (GA) enhanced with rule-based initialization to optimize mid-air refueling tasks. The method models refueling operations using spatial-temporal constraints, fuel-dependent task urgency, and mission priority weights. Unlike conventional GAs, our approach integrates domain knowledge by initializing the population with time-sorted feasible solutions and spatial feasibility checks, improving convergence speed and solution quality. A dynamic reward-penalty mechanism further prioritizes high-urgency refueling tasks. Simulation results across several scenarios demonstrate that the proposed algorithm outperforms random-initialization GAs, PSO, HGSA in paper[6], achieving faster convergence and more efficient scheduling. The framework shows strong potential for real-world applications, with future extensions targeting multi-RUAV cooperation and dynamic mission replanning.