Multi-Agent Collaborative Data-Engineering-Based Early Warning for Road Icing in High-Altitude Cold Mountainous Regions

doi:10.11871/jfdc.issn.2096-742X.2026.02.004

Abstract

Abstract:

[Objective] This study focuses on the limitations of traditional road icing warning methods in high-altitude mountainous environments, including poor adaptability, low coordination efficiency, and difficulties in practical implementation. It aims to explore and validate a new warning approach that integrates multi-agent collaboration with data engineering, with the goal of improving the accuracy, timeliness, and deployability of road icing warnings in such regions and providing a practical solution for engineering applications. [Methods] A new data engineering paradigm is established through the integration of large-scale artificial intelligence models, leveraging their advanced capabilities in semantic understanding, multimodal fusion, and adaptive reasoning to reconstruct the data engineering value chain. A standardized and automated “data-method-analysis-decision” intelligent pipeline is developed to enable end-to-end collaboration, encompassing data selection, engineering analysis, and decision support. This approach transforms data analysis from a one-off task into a continuously scalable and reusable engineering process. [Conclusions] This study proposes a fusion approach combining intelligent agents with data engineering and applies it to a road icing warning simulation on the G217 Duku Highway. By continuously analyzing video data from cameras and integrating multi-source heterogeneous data from along the route, this approach improves the accuracy and timeliness of road icing warnings while enhancing data reusability. The results indicate that the method is feasible for practical implementation in high-altitude, complex environments. It provides an efficient and reliable solution for road icing disaster warning and effectively promotes the transformation of data from a “resource” into a valuable “asset.”

Key words: data engineering, road icing warning, intelligent agents, AI, Duku Highway

LIU Jingqi, ZHANG Yaonan, KANG Jianfang, LIU Jie, YANG Zhiwei, ZHANG Zhixing, WANG Baode. Multi-Agent Collaborative Data-Engineering-Based Early Warning for Road Icing in High-Altitude Cold Mountainous Regions[J]. Frontiers of Data and Computing, 2026, 8(2): 40-53, https://cstr.cn/32002.14.jfdc.CN10-1649/TP.2026.02.004.

Figures/Tables 5

Fig.1

Fig.2

Fig.3

Fig.4

Table 1

References 62

[1]	张耀南. 数据工程学建设思考与实践[J]. 数据与计算发展前沿, 2022, 4(1): 5-19.
[2]	陈占龙, 陶留锋, 孙政权, 等. 数据工程视角下的军事地质数据体系研究[J]. 测绘与空间地理信息, 2022(11): 1-4.
[3]	张耀南, 田琛琛, 任彦润, 等. 自然灾害应急响应科学数据工程体系建设[J]. 数据与计算发展前沿, 2024, 6(1): 46-56.
[4]	JIN A A, YE Y, LEE B, et al. Decoagent: Large language model empowered decentralized autonomous collaboration agents based on smart contracts[J]. IEEE Access, 2024, 12: 155234-155245. doi: 10.1109/ACCESS.2024.3481641
[5]	HONG S, ZHUGE M, CHEN J, et al. MetaGPT: Meta Programming for Multi-Agent Collaborative Framework[C]. The 12th International Conference on Learning Representations, Vienna, Austria, 2024.
[6]	WANG J Y, XU H Y, JIA H T, et al. Mobile-agent-v2: Mobile device operation assistant with effective navigation via multi-agent collaboration[J]. Advances in Neural Information Processing Systems, 2024, 37: 2686-2710.
[7]	王宁. 道路路面缺陷检测算法与养护策略研究[D]. 山东交通学院, 2024.
[8]	杨二贺. 面向组合式图像检索的多模态特征对齐与融合方法研究[D]. 东华大学, 2025.
[9]	贾兴利, 曲远海, 朱浩然, 等. 交通预测中的时空图神经网络研究综述: 从模型解构到发展路径[J/OL]. 交通运输工程学报, 1-35[2025-10-25]. https://doi.org/10.19818/j.cnki.1671-1637.2026.01.003.
[10]	SHAO J. Fuzzy categorization of weather conditions for thermal mapping[J]. Journal of Applied Meteorology and Climatology, 2000, 39(10): 1784-1790.
[11]	VARGAS J, ALSWEISS S, TOKER O, et al. An overview of autonomous vehicles sensors and their vulnerability to weather conditions[J]. Sensors, 2021, 21(16): 5397. doi: 10.3390/s21165397
[12]	FLATSCHER M, NEUMAYER M, BRETTERLIEBER T, et al. Measurement of complex dielectric material properties of ice using electrical impedance spectroscopy[C]. IEEE Sensors, Orlando, USA, 2016: 1-3.
[13]	TROIANO A, NADDEO F, SOSSO E, et al. Assessment of force and fatigue in isometric contractions of the upper trapezius muscle by surface EMG signal and perceived exertion scale[J]. Gait & Posture, 2008, 28(2): 179-186. doi: 10.1016/j.gaitpost.2008.04.002
[14]	AMOIROPOULOS K, KIOSELAKI G, KOURKOUMELIS N, et al. Shaping beam profiles using plastic optical fiber tapers with application to ice sensors[J]. Sensors, 2020, 20: 2503. doi: 10.3390/s20092503
[15]	SIEGL A, NEUMAYER M, BRETTERLIEBER T. Fibre optical ice sensing: Sensor model and icing experiments for different ice types[C]. IEEE International Instrumentation and Measurement Technology Conference, Dubrovnik, Croatia, 2020: 1-6.
[16]	LI X, SHIIH W Y, VARTULI J, et al. Detection of water-ice transition using a lead zirconate titanate/brass transducer[J]. Journal of Applied Physics, 2002, 92: 106-111. doi: 10.1063/1.1481191
[17]	GU H, LI B, ZHANG X, et al. Detection of road surface water and ice based on polarization measurement[J]. Electronic Measurement Technology, 2011, 34: 99-102.
[18]	HORITA Y, SHIBATA K, MAEDA K, et al. Omni-directional polarization image sensor based on an omni-directional camera and a polarization filter[C]. IEEE International Conference on Advanced Video and Signal Based Surveillance, Genova, Italy, 2009: 280-285.
[19]	CASSELGREN J, SJÖDAHL M. Polarization resolved classification of winter road condition in the near-infrared region[J]. Applied Optics, 2012, 51: 3036-3045. doi: 10.1364/AO.51.003036
[20]	SUN Z, ZHANG J, ZHAO Y. Laboratory studies of polarized light reflection from sea ice and lake ice in visible and near infrared[J]. IEEE Geoscience and Remote Sensing Letters, 2012, 10: 170-173. doi: 10.1109/LGRS.2012.2196753
[21]	JONSSON P, CASSELGREN J, THÖRNBERG B. Road surface status classification using spectral analysis of NIR camera images[J]. Sensors, 2014, 15: 1641-1656.
[22]	JONSSON P. Remote sensor for winter road surface status detection[C]. IEEE Sensors, Limerick, Ireland, 2011: 1285-1288.
[23]	LIU J Q, ZHANG Y N, LIU J, et al. Automated Recognition of Snow-Covered and Icy Road Surfaces Based on T-Net of Mount Tianshan[J]. Remote Sensing, 2024, 16: 3727. doi: 10.3390/rs16193727
[24]	LEE S, JEON J, LE T H M. Feasibility of automated black ice segmentation in various climate conditions using deep learning[J]. Buildings, 2023, 13(3): 767. doi: 10.3390/buildings13030767
[25]	YUAN Y C, ISLAM M S, YUAN Y L, et al. EcRD: Edge-cloud computing framework for smart road damage detection and warning[J]. IEEE Internet of Things Journal, 2020, 8(16): 12734-12747. doi: 10.1109/JIOT.2020.3024885
[26]	郭晓澎. 主动式路面积雪凝冰监测预警及处置系统研究[J]. 现代信息科技, 2024, 8(21): 175-180.
[27]	SARABANDI K, LI E S, NASHASHIBI A. Modeling and measurements of scattering from road surfaces at millimeter-wave frequencies[J]. IEEE Transactions on Antennas and Propagation, 2002, 45(11): 1679-1688. doi: 10.1109/8.650080
[28]	CARSON J, MANNEERING F. The effect of ice warning signs on ice-accident frequencies and severities[J]. Accident Analysis & Prevention, 2001, 33(1): 99-109. doi: 10.1016/S0001-4575(00)00020-8
[29]	KIRCHER K, THORSLUND B. Effects of road surface appearance and low friction warning systems on driver behaviour and confidence in the warning system[J]. Ergonomics, 2009, 52(2): 165-176. doi: 10.1080/00140130802277547 pmid: 18937091
[30]	ZHOU Z, HUANG M H, HE J P, et al. Ice structure monitoring with an optical fiber sensing system[J]. Cold Regions Science and Technology, 2010, 61(1): 1-5. doi: 10.1016/j.coldregions.2009.12.004
[31]	TAN B. Road surface temperature monitor system realization based on Rich Internet Application model and GPRS technology[C]. The 2nd International Workshop on Intelligent Systems and Applications, Wuhan, China, 2010: 1-4.
[32]	PAPADIMITRATOS P, BUTTYAN L, HOLCZER T, et al. Secure vehicular communication systems: design and architecture[J]. IEEE Communications Magazine, 2008, 46(11): 100-109.
[33]	TANG J. Winter Road Weather Monitoring and Warning Service System[C]. The 11th International Conference of Chinese Transportation Professionals, Nanjing, China, 2011: 2080-2085.
[34]	CHENG Y C, He K. A New Monitoring and Early-Warning System on Large Range Road Using IPv6 Network[J]. Applied Mechanics and Materials, 2013, 253: 1135-1138.
[35]	LIU T, PAN Q, SANCHEZ J, et al. Prototype decision support system for black ice detection and road closure control[J]. IEEE Intelligent Transportation Systems Magazine, 2017, 9(2): 91-102.
[36]	DILORENZO T, YU X. Use of ice detection sensors for improving winter road safety[J]. Accident Analysis & Prevention, 2023, 191: 107197. doi: 10.1016/j.aap.2023.107197
[37]	LIU W J, CHEN Y Q, HOU F J, et al. Fixed spray type anti-ice deicing control system design[J]. Journal of Physics: Conference Series, 2019, 1423(1): 012001. doi: 10.1088/1742-6596/1423/1/012001
[38]	LIU W J, ZHANG Z P, LI D Z, et al. Design and Implementation of Intelligent Road Ice Early Warning System[C]. The 20th COTA International Conference of Transportation Professionals, 2020: 1955-1967.
[39]	CHARLTON S G. Conspicuity, memorability, comprehension, and priming in road hazard warning signs[J]. Accident Analysis & Prevention, 2006, 38(3): 496-506. doi: 10.1016/j.aap.2005.11.007
[40]	DING S, DONG X, ZHANG J, et al. Multi-Agent policy gradients with dynamic weighted value decomposition[J]. Pattern Recognition, 2025, 164: 111576. doi: 10.1016/j.patcog.2025.111576
[41]	ASGHARI S M, OUYANG Y, NAYYAR A. Regret bounds for decentralized learning in cooperative multi-agent dynamical systems[C]. Conference on Uncertainty in Artificial Intelligence, Toronto, Canada, 2020: 121-130.
[42]	HAN Z H, WANG J X, YAN X L, et al. CoReaAgents: A Collaboration and Reasoning Framework Based on LLM-Powered Agents for Complex Reasoning Tasks[J]. Applied Sciences, 2025, 15(10): 5663. doi: 10.3390/app15105663
[43]	DING Z, HUANG T, LU Z. Learning individually inferred communication for multi-agent cooperation[J]. Advances in Neural Information Processing Systems, 2020, 33: 22069-22079.
[44]	ZHANG K, YANG Z, BAŞAR T. Multi-agent reinforcement learning: A selective overview of theories and algorithms[J]. Handbook of reinforcement learning and control, 2021: 321-384.
[45]	YUAN Y, WENG X S, OU Y L, et al. Agentformer: Agent-aware transformers for socio-temporal multi-agent forecasting[C]. IEEE/CVF International Conference on Computer Vision Workshops, Montreal, Canada, 2021: 1-7.
[46]	YANG W K, BI X H, LIN Y K, et al. Watch out for your agents! Investigating backdoor threats to LLM-based agents[C]. Advances in Neural Information Processing Systems, 2024, 37: 100938-100964.
[47]	KOUVAROS P. Towards Formal Verification of Neuro-symbolic Multi-agent Systems[C]. The 32nd International Joint Conference on Artificial Intelligence, Macao, China, 2023: 7014-7019.
[48]	XU X W, ZHANG D W, LIU Q, et al. Agentic RAG with Human-in-the-Retrieval[C]. IEEE 22nd International Conference on Software Architecture Companion, Odense, Denmark, 2025: 498-502.
[49]	LUO Z R, LI D, WAN J F, et al. Multi-agent collaboration mechanisms based on distributed online meta-learning for mass personalization[J]. Journal of Industrial Information Integration, 2025, 46: 100852. doi: 10.1016/j.jii.2025.100852
[50]	TELNOV Y F, KAZAKOV V A, DANILOV A V. Designing a multi-agent system for a network enterprise[J]. Бизнес-информатика, 2024, 18(3): 70-86.
[51]	DAS A, CHEN S C, SHYU M L, et al. Enabling synergistic knowledge sharing and reasoning in large language models with collaborative multi-agents[C]. IEEE 9th International Conference on Collaboration and Internet Computing, Atlanta, GA, USA, 2023: 92-98.
[52]	GAO P Y, ZHAO J M, CHEN X Y, et al. An Efficient Context-Dependent Memory Framework for LLM-Centric Agents[C]. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics, Albuquerque, New Mexico, 2025: 1055-1069.
[53]	HOU X Y, FORSYTH C, ANDREWS-TODD J, et al. An LLM-Enhanced Multi-agent Architecture for Conversation-Based Assessment[C]. International Conference on Artificial Intelligence in Education, Palermo, Italy, 2025: 119-134.
[54]	TOPSAKAL O, AKINCI T C. Creating large language model applications utilizing langchain: A primer on developing llm apps fast[C]. International Conference on Applied Engineering and Natural Sciences, Konya, Turkey, 2023, 1(1): 1050-1056.
[55]	WAN J B, ZHANG F, PAN J F. Promoting organized basic research: Strategic layout and strategic capacity in science and technology[J]. Bulletin of Chinese Academy of Sciences (Chinese Version), 2021, 36(12): 1404-1412.
[56]	肖楚琬, 刘嘉, 唐小峰. 面向智能决策的军事数据工程建设[J]. 指挥控制与仿真, 2020(3): 97-101. doi: 10.3969/j.issn.1673-3819.2020.03.018
[57]	畅婷婷. 匈塞铁路联锁数据工程化生成方法与实现[J]. 铁路通信信号工程技术, 2023(3): 91-96.
[58]	李少芳, 张家成, 李伟. 装备试验中的数据流程重构研究[J]. 兵器装备工程学报, 2022, 43(12): 99-104.
[59]	周欣欣, 涂积奇, 王常月, 等. 航天型号数据工程应用实践[J]. 中国质量与标准导报, 2023(3): 62-65.
[60]	CHUN D X, JUN C Y, ZHONG C Y, et al. Data Engineering in Information System Construction[C]. Proceedings of the International Symposium on Room Acoustics, Kuala Lumpur, Malaysia, 2012: 135-137.
[61]	BAESENS B, HÖPPNER S, VERDONCK T. Data Engineering for Fraud Detection[J]. Decision Support Systems, 2021, 140: 113429. doi: 10.1016/j.dss.2020.113429
[62]	马跃飞. 数据工程标准体系框架研究现状与建议[J]. 电子技术与软件工程, 2019(9): 183-184.

序号	摄像头	经度	纬度	路面状况	气温（℃）	风速（m/s）	降水（m）
1	G217奎屯地区哈希勒根防雪走廊G217-K668+966	84°25′14.52″	43°43′59.47″	积雪路面	nan	0.8	0
2	G217伊犁地区乔尔玛烈士陵园G217-K694+800	84°21′32.55″	43°39′8.68″	积雪路面	-7.1	0.8	0
3	G217伊犁地区巴音布鲁克草原G217-K817+000	84°19′32.32″	43°4′22.47″	积雪路面	-10	6.9	0
4	G217阿克苏地区铁力买提隧道G217-K931+183	83°26′0.97″	42°28′41.27″	积雪路面	-5.6	6.3	0.9
5	G217阿克苏地区库车市大龙池G217-K955+000	83°16′38.51″	42°27′4.21″	积雪路面	nan	6.3	0.9
6	G217阿克苏地区库车市盐水沟隧道东G217-K1067+070	82°49′7.16″	41°51′22.96″	干燥路面	nan	0.8	3.1