Volume 2,Issue 7
基于图神经网络和多目标粒子群优化的智能路径规划方法及其应用
随着智能交通以及旅游规划的不断发展，对有限时间、预算的约束条件下优化旅行路线、提高赏花效益的问题越来越突出。为了克服与假日赏花旅游有关的城市访问和路线规划难题，本文提出了将图神经网络（GNN）和多目标粒子群优化（MOPSO）结合起来的智能路线规划方法（GNN-MOPSO）。该方法用GNN 从城市网络中提取结构特征，用MOPSO 优化多个目标，从而达到同时最大化赏花价值、最小化旅行成本的目的。从实验结果可以看出，在处理大规模城市网络的时候，GNN-MOPSO 比传统的PSO、NSGA-II 等方法在解决方案质量和多样性上要好。另外GNNMOPSO算法可以很好地处理复杂的约束，具有在智能交通、旅游路线优化等方面的应用前景。最后，本文对未来的几个研究方向进行了论述，即强化学习的集成以及算法在更大的城市网络中是否可以推广。
[1]McNulty A ,Berman O B ,Engelbrecht A .A comparative study of evolutionary algorithms and particle swarm optimization approaches for constrained multi-objective optimization problems[J].Swarm and Evolutionary Computation,2024,91101742-101742.DOI:10.1016/J.SWEVO.2024.101742.
[2]Dezvarei M ,Tomsovic K ,Sun S J , et al.A graph neural network framework for security assessment using topological measures[J].Electric Power Systems Research,2025,249111972-111972.DOI:10.1016/J.EPSR.2025.111972.
[3]Kajla I N ,Missen S M M ,Coustaty M , et al.A histogram-based approach to calculate graph similarity using graph neural networks[J].Pattern Recognition Letters,2024,186286-291.DOI:10.1016/J.PATREC.2024.10.015.
[4]Zhao Y ,Dong J ,Wang W , et al.A multi-typed multi-relational heterogeneous graph neural network model for complex networks[J].Knowledge-Based Systems,2025,329( PA):114291-114291.DOI:10.1016/J.KNOSYS.2025.114291.
[5]Quintero H P ,Triviño G P ,González C D , et al.Adaptive multi-objective real-time hierarchical control for isolated microgrid clusters utilizing an enhanced particle swarm optimization strategy to optimize costs and emissions[J].Electric Power Systems Research,2026,250112169-112169.DOI:10.1016/J.EPSR.2025.112169.
[6]Meng X ,Li H .An adaptive co-evolutionary competitive particle swarm optimizer for constrained multi-objective optimization problems[J].Swarm and Evolutionary Computa tion,2024,91101746-101746.DOI:10.1016/J.SWEVO.2024.101746.
[7]Cunha B C ,Massarotto F D ,Fornazza L S , et al.An ALNS metaheuristic for the family multiple traveling salesman problem[J].Computers and Operations Research,2024,169106750-106750.DOI:10.1016/J.COR.2024.106750.
[8]Luo X ,Yi Z .Efficiency management of engineering projects based on particle swarm multi objective optimization algorithm[J].Systems and Soft Computing,2025,7200320-200320.DOI:10.1016/J.SASC.2025.200320.
[9]Yang X ,Wang R ,Li K , et al.Exploratory landscape analysis on black-box optimization problems via Graph Neural Network[J].Swarm and Evolutionary Computation,2025,99102136-102136.DOI:10.1016/J.SWEVO.2025.102136.
[10]Wu J ,Xu Z ,Qiao S .Fast Heterogeneous Graph Neural Network Generation via Meta Contrastive Learning.[J].Neural networks : the official journal of the International Neural Network Society,2025,192107727.DOI:10.1016/J.NEUNET.2025.107727.
[11]Wang X ,Lv Y ,Sun H , et al.Multi-modal travel route planning considering environmental preference under uncertainties: A distributionally robust optimization approach[J].Transportation Research Part E,2025,198104097-104097.DOI:10.1016/J.TRE.2025.104097.
[12]Zhao Y ,Wang W ,Wang S , et al.Over-smoothing problem of heterogeneous graph neural networks: A heterogeneous graph neural network with enhanced node differentiability[J].Information Processing and Management,2026,63(2PA):104395-104395.DOI:10.1016/J.IPM.2025.104395.
[13]Dinçer H ,Yüksel S ,Eti S , et al.Q-learning algorithm and molecular fuzzy multi-objective particle swarm optimization-based decision-making approach to circular economy-oriented investment alternatives for renewable energy technologies[J].Information Sciences, 2025,718122378-122378.DOI:10.1016/J.INS.2025.122378.
[14]Lin C ,Xie Y ,Wang H C .Significant wave height prediction at multiple sites using sequence decomposition and dynamic spatiotemporal graph neural networks[J].Ocean Engineering,2025,341(P2):122548-122548.DOI:10.1016/J.OCEANENG.2025.122548.
[15]Feng B ,Zhou P X .The novel physics-enhanced graph neural network for phase-field fracture modelling[J].Computer Methods in Applied Mechanics and Engineering,2025,446(PB):118284-118284.DOI:10.1016/J.CMA.2025.118284.