Resilient Low-Altitude Logistics: A Multi-Objective Evolutionary Approach for UAV Network Topology Optimization
Please login to view abstract download link
The rapid expansion of the low-altitude economy presents unprecedented opportunities for logistics but demands highly resilient network architectures. Designing Unmanned Aerial Vehicle (UAV) networks that balance operational efficiency, structural robustness, and construction costs is a complex multi-objective challenge. This paper proposes an AI-driven optimization framework based on the Non-dominated Sorting Genetic Algorithm II (NSGA-II). We introduce a hybrid topology generation mechanism that integrates Minimum Spanning Trees (MST) for backbone connectivity with Voronoi-based partitioning for access networks. The method optimizes three conflicting objectives: global transport efficiency, network robustness (measured by algebraic connectivity), and operational cost. Numerical experiments confirm the algorithm’s effectiveness, with the Hypervolume indicator proving mathematical convergence to a diverse set of Pareto-optimal solutions. The results identify critical trade-offs between cost and resilience, providing a scalable, data-driven methodology for the strategic planning of future urban air mobility ecosystems [1, 2, 3].
