Hybrid Approach for Multi-Robot Task Allocation in Search and Rescue Operations Using NSGA-II and Variable Neighborhood Descent

This work presents a hybrid optimization approach combining the global search capabilities of the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) with Variable Neighborhood Descent (VND). The proposed methodology addresses critical SAR operations objectives, such as minimizing mission completion time, prioritizing high-impact tasks, and balancing UAV energy consumption. NSGA-II provides a robust framework for exploring the solution space and generating diverse Pareto-optimal solutions, while VND systematically refines them for better convergence and quality. Simulation results indicate statistically significant improvements, achieving average reductions of 4.80% in travel distance and 8.91% in mission completion time compared to standalone NSGA-II. Although the current framework does not support dynamic task reassignment in real-time, it demonstrates potential for practical applications in static SAR scenarios.