大模型在交通运输行业的应用路径

张丽; 郑晓峰; 刘礼勇; 孙逸帆; 林垚

doi:10.16503/j.cnki.2095-9931.2025.04.006

交通运输研究 ›› 2025, Vol. 11 ›› Issue (4) : 67-78. DOI: 10.16503/j.cnki.2095-9931.2025.04.006

专刊：交通运输数字化转型

大模型在交通运输行业的应用路径

张丽 ¹ ,
郑晓峰 ²^,^* ,
刘礼勇 ¹ ,
孙逸帆 ¹ ,
林垚 ¹

作者信息 +

Application Paths of Large Models in Transportation Industry

ZHANG Li ¹ ,
ZHENG Xiaofeng ²^,^* ,
LIU Liyong ¹ ,
SUN Yifan ¹ ,
LIN Yao ¹

Author information +

文章历史 +

摘要

为加快推动人工智能与交通运输的深度融合，通过人工智能技术赋能交通运输高质量发展，以交通运输典型场景带动人工智能技术迭代升级，研究提出大模型在交通运输行业的应用路径。首先，梳理大模型在交通运输行业应用研究进展，总结典型应用场景，分析应用研究特征和存在的主要问题。其次，开展大模型在交通运输行业的技术适配性研究，解构大模型的核心技术能力，分析交通建设、管理、养护、运营和服务五大领域的痛点和大模型支撑解决行业痛点问题的能力，对大模型在交通运输行业的应用进行SWOT分析。最后，研究提出包括总体架构、三类重点应用场景、三种技术路径和实施策略的大模型应用路径。研究结果可为大模型在交通运输行业的应用落地提供理论和方法参考。

Abstract

To accelerate the deep integration of artificial intelligence and transportation, empower high-quality development of transportation with artificial intelligence technology, and drive the iterative upgrading of artificial intelligence technology with typical transportation scenarios, the paper studies and proposes the application paths of large models in the transportation industry. Firstly, it reviews the research progress of the application of large models in the transportation industry, summarizes typical application scenarios, and analyzes the characteristics and main problems. Secondly, it conducts research on the technological adapatability of large models in the transportation industry. It deconstructs the core technical capabilities of large models, analyzes the pain points in the five major fields of transportation including construction, management, maintenance, operation, and service, and discusses the ability of large models to support the resolution of industry pain points. It conducts a SWOT analysis on the application of large models in the transportation industry. Finally, the application paths of the large models including the overall architecture, three key application scenarios, three technical paths, and implementation strategies were proposed. The research results can provide theoretical and methodological reference for the application and implementation of large-scale models in the transportation industry.

导出引用

张丽, 郑晓峰, 刘礼勇, 等. 大模型在交通运输行业的应用路径[J]. 交通运输研究. 2025, 11(4): 67-78 https://doi.org/10.16503/j.cnki.2095-9931.2025.04.006

ZHANG Li, ZHENG Xiaofeng, LIU Liyong, et al. Application Paths of Large Models in Transportation Industry[J]. Transport Research. 2025, 11(4): 67-78 https://doi.org/10.16503/j.cnki.2095-9931.2025.04.006

中图分类号： U495

参考文献

列表( 原文顺序 | 文献年度倒序 | 文中引用次数倒序 ) 可视化分析

[1]	TU X, HE Z, HUANG Y, et al. An overview of large AI models and their applications[J]. Visual Intelligence, 2024, 2(1): 34.

[2]	LIU W, WANG Z, LIU X, et al. A survey of deep neural network architectures and their applications[J]. Neurocomputing, 2016, 23(4): 11-26.

[3]	YIN S, FU C, ZHAO S, et al. A survey on multimodal large language models[J]. National Science Review, 2024, 11(12). DOI:10.1093/nsr/nwae403.

[4]	黄世泽, 谭群垚, 吴麒骥, 等. 交通科普大模型的构建方法研究[J]. 上海公路, 2024(4):217-222,227,244.

[5]	WANG X, LYU J, SLOWIK A, et al. Large models for cooperative control of connected and autonomous vehicles[J]. IEEE Transactions on Vehicular Technology, 2025, 74(2): 1935-1948.

[6]	邱暾, 孙亮, 付本贺, 等. 大模型在智慧高速公路领域中的应用探索[J]. 中国交通信息化, 2024(S1):38-41.

[7]	王辉, 李金明, 张丽, 等. 以人工智能驱动的新质生产力赋能交通运输高质量发展的思考与对策建议[J]. 交通运输研究, 2024, 10(2):11-19.

[8]	WANG P, WEI X, HU F, et al. TransGPT:Multi-modal generative pre-trained transformer for transportation[C]// Proceedings of 2024 International Conference on Computational Linguistics and Natural Language Processing ( CLNLP). Yinchuan,China: IEEE, 2024: 96-100.

[9]	ZHANG S, FU D, LIANG W, et al. TrafficGPT: Viewing, processing and interacting with traffic foundation models[J]. Transport Policy, 2024, 150:95-105.

[10]	YUAN Y, DING J, FENG J, et al. UniST: a prompt-empowered universal model for urban spatiotemporal prediction[C]// Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York: ACM, 2024: 4095-4106.

[11]	HU Y, YANG J, CHEN L, et al. Planning-oriented autonomous driving[C]// Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR). Vancouver, Canada: IEEE, 2023: 17853-17862.

[12]	武卫红. 佳都科技董事长刘伟:推动大模型与大交通“知行共进”[N]. 中国证券报,2023-07-01(A04).

[13]	赵雯琪. “钻研ChatGPT”一年多后顺丰推出“丰知”物流决策大模型[N]. 每日经济新闻,2024-08-21(008).

[14]	YAO H, DA L, NANDAM V, et al. CoMAL:Collaborative multi-agent large language models for mixed-autonomy traffic[C]// Proceedings of the 2025 SIAM International Conference on Data Mining ( SDM). Alexandria USA: SIAM, 2024: 409-418.

[15]	周臻, 顾子渊, 曲小波, 等. 城市多模式交通大模型MT-GPT:点线面的分层技术与应用场景[J]. 中国公路学报, 2024, 37(2):253-274.

[16]	肖建力, 邱雪, 张扬, 等. 交通大模型综述[J]. 交通运输工程学报, 2025, 25(1):8-28.

[17]	ZHAO Y, HU C, ZHU Z, et al. Crowd sensing intelligence for ITS: Participants, methods, and stages[J]. IEEE Transactions on Intelligent Vehicles, 2023, 8: 3541-3546.

[18]	DA L, GAO M, MEI H, et al. LLM Powered sim-to-real transfer for traffic signal control[J]. arXiv Preprint. arXiv: 2308.14284. DOI: 10.48550/arXiv.2308.14284.

[19]	MICHAEL V, BIBEK P, LI W. Can ChatGPT enable ITS? The case of mixed traffic control via reinforcement learning[C]// Proceedings of 2023 IEEE 26th International Conference on Intelligent Transportation Systems(ITSC). Bilbao, Spain:IEEE, 2023: 3749-3755.

[20]	吴精乙, 景峻, 贺熠凡, 等. 基于多模态大模型的高速公路场景交通异常事件分析方法[J]. 图学学报, 2024, 45(6):1266-1276.

[21]	ZHOU X, LIU M, EKIM Y, et al. Vision language models in autonomous driving and intelligent transportation systems[EB/OL]. (2023-10-22)[2025-06-12]. https://doi.org/10.48550/arXiv.2310.14414.

[22]	LEI L, ZHANG H, YANG S, ChatGPT in connected and autonomous vehicles: Benefits and challenges[J]. Intell Robot, 2023, 3: 145-148.

[23]	ZHAO Y, LI Z, JIN Z, et al. Enhancing the spatial awareness capability of multi-modal large language model[J]. arXiv Preprint. arXiv: 2310.20357. DOI: 10.48550arXiv.2310.20357.

[24]	CUI C, MA Y, CAO X, et al. A survey on multimodal large language models for autonomous driving[C]// Proceedings of 2024 IEEE/CVF Winter Conference on Applications of Computer Vision. Waikoloa, USA: IEEE, 2024: 958-979.

[25]	金彦亮, 顾晨杰, 高塬. 基于多模态大语言模型的低延迟端到端自动驾驶模型[J]. 工业控制计算机, 2025, 38(3):32-34.

[26]	CUI Y, HUANG S, ZHONG J, et al. DriveLLM: Charting the path toward full autonomous driving with large language models[J]. IEEE Transactions on Intelligent Vehicles, 2023, 9: 1450-1464.

[27]	HAO S, MU Y, JIANG Y, et al. LanguageMPC: large language models as decision makers for autonomous driving[C]// Proceedings of The Twelfth International Conference on Learning Representations(ICLR).Vienna, Austria: ICLR, 2024.

[28]	DENG Y, TU Z, YAO J, et al. TARGET: Traffic rule-based test generation for autonomous driving systems[J]. IEEE Transactions on Software Engineering, 2025: 1-19.

[29]

RAJA

K V

, SIDDHARTH

, YUVARAJ

, et al. An Artificial Intelligence based automated case-based reasoning (CBR) system for severity investigation and root-cause analysis of road accidents - Comparative analysis with the predictions of ChatGPT[J]. Journal of Eengineering Research, 2024, 12(4): 895-903.

[30]	DE ZARZÀ, DE CURTÒ, ROIG G, et al. LLM multimodal traffic accident forecasting[J]. Sensors, 2023, 23. DOI: 10.3390/S23229225.

[31]	ARTEAGA C, PARK J. A large language model framework to uncover underreporting in traffic cra-shes[J]. Journal of Safety Research, 2025, 9: 1-13.

[32]	王文浩. 城市轨道交通系统风险链群模型构建与应用方法研究[D]. 北京: 北京交通大学, 2024.

[33]	DONG C, SYED N, JIANG F, et al. Securing UAV delivery systems with blockchain and large language models: An innovative logistics solution[C]// Proceedings of 2024 11th International Conference on Machine Intelligence Theory and Applications. Melbourne, Australia: IEEE, 2024: 42-49.

[34]	VOSS S. Successfully using ChatGPT in logistics: Are we there yet?[C]// Proceedings of Computational Logistics: 14th International Conference, ICCL 2023. Berlin, Germany: Springer, 2023: 3-17.

[35]	JACKSON I, SAENZ M, IVANOV D. From natural language to simulations: Applying AI to automate simulation modelling of logistics systems[J]. International Journal of Production Research, 2024, 62(4): 1434-1457.

[36]	KMIECIK M. ChatGPT in third-party logistics - The game-changer or a step into the unknown?[J]. Journal of Open Innovation: Technology, Market, and Complexity, 2023, 9(4). DOI:10.1016/j.joitmc.2023.100174.

[37]	肖梅宁. 智能物流进化:人工智能大模型重塑供应链生态和智能化未来[J]. 物流科技, 2024, 47(17):130-132.

[38]	CHEN X, ZHANG L. Revolutionizing bridge operation and maintenance with LLM-based agents: an overview of applications and insights[J]. arXiv Preprint. arXiv: 2407.10064. DOI: 10.48550/arXiv.2407.10064.

[39]	刘晨光, 金峻民, 王钰龙. 基于预训练大模型的智能运维系统平台构建[J]. 技术与市场, 2023, 30(9):13-18.

[40]	杨玉林, 胡伟, 贾晓, 等. 基于大模型技术的港口设备运维助手智能体系统[J]. 港口科技, 2025(2):1-5,15.

[41]	陈晨. 探索AI大模型在交通执法系统中的应用[J]. 人民公交, 2024(22):143-145.

[42]	夏宝山. 智能客服中心在轨道交通中的研究与应用[C]// 第三十七届中国(天津)2023′ IT、网络、信息技术、电子、仪器仪表创新学术会议论文集. 天津, 中国: 天津市电子学会,2023:96-98.

[43]	赵耀文, 曹振波. 基于功能系统理论的创新与价值创造方法论研究——兼论价值工程理论研究的前沿趋势[J]. 价值工程, 2019, 38(17):296-299.

[44]	程赛楠. 基于创新驱动发展的我国技术创新扩散理论文献综述[J]. 全国流通经济, 2019(21):124-125.

[45]	罗青军. 波特价值链模型的另画及其解释[J]. 商业研究, 2002(10):28-30.

[46]	钟升达, 康陈. AI大模型在公路交通领域的应用[EB/OL]. (2024-12-29)[2025-06-12]. https://mp.weixin.qq.com/s?__biz=MzUzMjc5ODY3NA==&mid=2247530450&idx=2&sn=0bdbdf0f3fa80566001fd523fe815154&chksm=faaffa6fcdd87379442c3b79fbff31aacd0aeefe3b0862305c735d56d9f1ba049f4bd4091650&scene=27.

[47]	BENZAGHTA M, ELWALDA A, MOUSA M, et al. SWOT analysis applications: An integrative literature review[J]. Journal of Global Business Insights, 2021, 6(1): 54-72.

[48]	金宏意. 基于深度强化学习的交叉口信号智能控制研究[D]. 重庆: 重庆交通大学, 2023.

[49]	肖婧, 王科俊, 毕晓君. 交叉口混合交通流高维多目标信号优化控制[J]. 公路交通科技, 2014, 31(11):108-115.

[50]	肖文妮. 面向智能交通的动态路径规划分析[J]. 九江学院学报(自然科学版), 2024, 39(2): 46-50.

[51]	杨璐. 基于移动GIS的城市交通动态路径规划及系统实现[D]. 大连: 辽宁师范大学, 2017.

[52]	孟聪. 基于多Agent的物流配送车辆智能调度研究[D]. 厦门: 集美大学, 2017.

[53]	李义华. 基于多智能体的物流配送车辆调度决策方法研究[D]. 长沙: 中南大学, 2012.

[54]	周熠洋. 基于大群体决策和多目标优化的沥青路面养护规划研究[D]. 武汉: 华中科技大学, 2024.

[55]	肖烽. 高速公路沥青路面养护管理与优化决策研究[D]. 南京: 东南大学, 2023.

[56]	GAO Y, XIONG Y, GAO X, et al. Retrieval-augmented generation for large language models: A survey[J]. arXiv Preprint. arXiv: 2312.10997. DOI: 10.48550/arXiv.2312.10997.