Similar articles

Abstract: Siting low-altitude takeoff and landing platforms (vertiports) is a fundamental challenge for developing urban air mobility (UAM). This study formulates this issue as a variant of the capacitated facility location problem, incorporating flight range and service capacity constraints, and proposes SPID, a deep reinforcement learning (DRL)-based solution framework that models the problem as a Markov decision process. To handle dynamic coverage, the designed DRL framework-based SPID uses a multi-head attention mechanism to capture spatiotemporal patterns, followed by integrating dynamic and static information into a unified input state vector. Afterward, a gated recurrent unit (GRU) is used to generate the query vector, thereby enhancing sequential decision-making. The action network within the DRL network is regulated by a loss function that integrates service distance costs with unmet demand penalties, enabling end-to-end optimization. Subsequent experimental results demonstrate that SPID significantly enhances solution efficiency and robustness compared with traditional methods under flight and capacity constraints. Especially, across the social performance metrics emphasized in this study, SPID outperforms the suboptimal solutions produced by traditional clustering and graph neural network (GNN)-based methods by up to approximately 29%. This improvement comes with an increase in distance-based cost that is kept within 10%. Overall, we demonstrate an efficient, scalable approach for vertiport siting, supporting rapid decision-making in large-scale UAM scenarios.

SPID：一个基于深度强化学习的低空起降设施选址解决方案框架

[1]Abdel-Basset M, Mohamed R, Hezam IM, et al., 2024. Multiobjective trajectory optimization algorithms for solving multi-UAV-assisted mobile edge computing problem. J Cloud Computing, 13:35.

[2]Apaza RD, Li HX, Han RX, et al., 2023. Multi-agent deep reinforcement learning for spectrum and air traffic management in UAM with resource constraints. IEEE/AIAA 42^nd Digital Avionics Systems Conf, p.1-7.

[3]Berger R, 2023. Urban Air Mobility Market Update. https://www.rolandberger.com [Accessed on June 30, 2025].

[4]Church R, ReVelle C, 1974. The maximal covering location problem. Papers Region Sci, 32:101-118.

[5]Deloitte, 2023. Advanced Air Mobility Industry Insight. https://www2.deloitte.com [Accessed on June 28, 2025].

[6]Escribano Macias J, Khalife C, Slim J, et al., 2023. An integrated vertiport placement model considering vehicle sizing and queuing: a case study in London. J Air Transp Manage, 113:102486.

[7]Federal Aviation Administration, 2023. Aerospace Forecast Fiscal Years 2023–2043. https://www.faa.gov/data_research/aviation/aerospace_forecasts/faa-aerospace-forecast-fy-2023-2043 [Accessed on June 28, 2025].

[8]Garcia CP, Li WG, Hirata NST, et al., 2023. ISUAM: intelligent and safe UAM with deep reinforcement learning. IEEE 29^th Int Conf on Parallel and Distributed Systems, p.378-383.

[9]Gopi SP, Magarini M, 2021. Reinforcement learning aided UAV base station location optimization for rate maximization. Electronics, 10(23):2953.

[10]Hakimi SL, 1965. Optimum distribution of switching centers in a communication network and some related graph theoretic problems. Oper Res, 13(3):462-475.

[11]Han RX, Li HX, Apaza R, et al., 2022. Deep reinforcement learning assisted spectrum management in cellular based urban air mobility. IEEE Wirel Commun, 29(6):14-21.

[12]Han XY, Xie MX, Yu K, et al., 2024. Combining graph neural network with deep reinforcement learning for resource allocation in computing force networks. Front Inform Technol Electron Eng, 25(5):701-712.

[13]Hottung A, Tierney K, 2020. Neural large neighborhood search for the capacitated vehicle routing problem. 24^th European Conf on Artificial Intelligence, p.443-450.

[14]Kim D, Lee K, Moon I, 2019. Stochastic facility location model for drones considering uncertain flight distance. Ann Oper Res, 283(1-2):1283-1302.

[15]Kool W, Van Hoof H, Welling M, 2019. Attention, learn to solve routing problems!

[16]Kumar PK, Witter J, Paul S, et al., 2023. Graph learning based decision support for multi-aircraft take-off and landing at urban air mobility vertiports. AIAA SciTech 2023 Forum.

[17]Lee S, Cho N, 2025. Optimal location of urban air mobility (UAM) vertiport using a three-stage geospatial analysis framework. Fut Transp, 5(2):58.

[18]Li M, Zhao P, Bao ZA, et al., 2025. A reinforcement learning-based iterative method for capacitated hub location problems in UAV networks. Integrated Communications, Navigation and Surveillance Conf, p.1-11.

[19]Liang HJ, Wang SH, Li HL, et al., 2024. SpoNet: solve spatial optimization problem using deep reinforcement learning for urban spatial decision analysis. Int J Digit Earth, 17(1):2299211.

[20]Lynskey J, Thar K, Oo TZ, et al., 2019. Facility location problem approach for distributed drones. Symmetry, 11(1):118.

[21]Mahmoodi A, Sajadi SM, Sadeq AM, et al., 2025. Enhancing unmanned aerial vehicles logistics for dynamic delivery: a hybrid non-dominated sorting genetic algorithm II with Bayesian belief networks. Ann Oper Res, early access.

[22]Meng ZY, Yu K, Qiu R, 2024. Location-routing optimization of UAV collaborative blood delivery vehicle distribution on complex roads. Complex Intell Syst, 10(6):8127-8141.

[23]Meskar M, Ahmadi-Javid A, 2024. Optimizing drone delivery paths from shared bases: a location-routing problem with realistic energy constraints. J Intell Robot Syst, 110(4):142.

[24]National Aeronautics and Space Administration, 2020. STEM LEARNING: Package Delivery Drone Simulation Coding Activity Guide. https://www.nasa.gov/wp-content/uploads/2020/05/aam-package-delivery-drone-simulation-activity-guide_0.pdf [Accessed on June 26, 2025].

[25]Nazari M, Oroojlooy A, Takáč M, et al., 2018. Reinforcement learning for solving the vehicle routing problem. Proc 32^nd Int Conf on Neural Information Processing Systems, p.9861-9871.

[26]Park C, Kim GS, Park S, et al., 2023. Multi-agent reinforcement learning for cooperative air transportation services in city-wide autonomous urban air mobility. IEEE Trans Intell Veh, 8(8):4016-4030.

[27]Paul A, Levin MW, Waller ST, et al., 2024. Data-driven optimization for drone delivery service planning with online demand.

[28]Premkumar VGR, Van Scoy B, 2025. Optimal positioning of unmanned aerial vehicle (UAV) base stations using mixed-integer linear programming. Drones, 9(1):44.

[29]Rahman MA, Basheer MA, Khalid Z, et al., 2024. Logistics hub location optimization: a k-means and p-median model hybrid approach using road network distances.

[30]Ryu K, Kim W, 2024. Energy efficient deployment of aerial base stations for mobile users in multi-hop UAV networks. Ad Hoc Netw, 157:103463.

[31]Shavarani SM, Nejad MG, Rismanchian F, et al., 2018. Application of hierarchical facility location problem for optimization of a drone delivery system: a case study of Amazon Prime Air in the city of San Francisco. Int J Adv Manuf Technol, 95(9):3141-3153.

[32]Shavarani SM, Mosallaeipour S, Golabi M, et al., 2019. A congested capacitated multi-level fuzzy facility location problem: an efficient drone delivery system. Comput Oper Res, 108:57-68.

[33]Song KH, Lee H, 2025. Network topology-driven vertiport placement strategy: integrating urban air mobility with the Seoul Metropolitan railway system. Appl Sci, 15(7):3965.

[34]Sun J, Shu SJ, Hu HD, et al., 2025. Location optimization of unmanned aerial vehicle (UAV) drone port for coastal zone management: the case of Guangdong coastal zone in China. Ocean Coast Manag, 262:107576.

[35]Volakakis V, Mahmassani HS, 2024. Vertiport infrastructure location optimization for equitable access to urban air mobility. Infrastructures, 9(12):239.

[36]Williams RJ, 1992. Simple statistical gradient-following algorithms for connectionist reinforcement learning. Mach Learn, 8(3-4):229-256.

[37]Zhang CX, Du WB, Guo T, et al., 2025. Multi-objective hub location for urban air mobility via self-adaptive evolutionary algorithm. Adv Eng Inform, 64:102974.

[38]Zhao Y, Feng T, 2024. Strategic integration of vertiport planning in multimodal transportation for urban air mobility: a case study in Beijing, China. J Clean Prod, 467:142988.

[39]Zheng L, Xu G, Chen WB, 2024. Using improved particle swarm optimization algorithm for location problem of drone logistics hub. Comput Mater Con, 78(1):935-957.

[40]Zhu TK, Boyles SD, Unnikrishnan A, 2022. Two-stage robust facility location problem with drones. Transp Res Part C Emerg Technol, 137:103563.