Answers are grounded in this paper and cite their sources.
This paper investigates the planning of urban low-altitude public air route networks for drone logistics. Given the vertiports, their flight demand and the three-dimensional building environment, a route network is sought that serves all demand while minimizing both the demand-weighted flight cost and the total network length, the main determinant of deployment cost. We formulate this bi-objective problem and propose a two-stage framework. Stage 1 generates obstacle-free candidate routes for each origin–destination (OD) pair by layered A* search, with a collinearity-aware mechanism that draws the routes of different pairs onto shared corridors. Stage 2 selects among the candidates with NSGA-II, in which a connectivity repair operator keeps all OD pairs reachable. Unlike an exact joint formulation of path and network design, whose computational cost grows rapidly with instance size, the decomposition remains tractable at city scale. In ablations, the collinearity-aware mechanism raises the hypervolume (HV) by 21.25% over a single-shortest-path baseline on a compact instance (10 OD pairs) and shortens the network by 16–23% at equal flight cost on a city-scale Shenzhen instance (169 vertiports, 203 OD pairs), where the repair operator raises HV by 9.29% over NSGA-II without it.
No takes yet. Share an insight, caveat, or question.
Zhao et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: