| 85 | 0 | 338 |
| 下载次数 | 被引频次 | 阅读次数 |
【背景】物联网和智能设备技术的快速发展，为个体出行轨迹数据的监测提供了技术基础，但GPS等定位设备存在的原始定位误差，导致记录点位偏离实际位置，特别在城市密集路网中，无法精准还原真实的出行路径。【目标】提出一种路网约束下基于动态速度模拟的个体级轨迹重构方法，旨在解决离散GPS轨迹数据在城市道路网络中的连续映射问题，生成长时间、高精度的车辆或个体在路网中的连续移动轨迹。【方法】通过融合最短路径搜索、动态速度模型和时间驱动插值技术，构建包含空间映射-路径规划-速度采样-时间插值的四级重构机制。【数据】以上海市5万用户的手机信令数据为实验对象，并结合了上海市路网拓扑数据。【结论】轨迹重构结果显示：对比三类主流方法，本方法在平均轨迹误差方面分别提升33.3%、21.1%、12.5%，可精准刻画高精轨迹。轨迹数据能准确捕捉城市交通的时空特征，例如早晚高峰流量变化、虹桥枢纽等多中心区域的辐射模式等。进一步在北京和厦门的应用验证表明，该方法具有跨城市的适用性，可为交通流量预测、出行起讫点（OD）分析等智慧交通应用提供可靠的数据基础和技术支持。
Abstract：[Background] The rapid advancement in the Internet of Things(IoT) and smart-device technologies has enabled the monitoring of individual mobility trajectories. However, inherent positioning errors in GPS and other location-tracking devices often result in recorded points that deviate from their true locations, particularly in dense urban road networks, making it challenging to accurately reconstruct actual travel paths. [Objective] This study proposes an individual-level trajectoryreconstruction method based on dynamic velocity simulation and map matching to continuously map discrete global positioning system(GPS) trajectories in urban road networks. The goal is to generate long-term high-precision continuous-movement trajectories of vehicles or individuals that adhere to road-network constraints. [Method] The proposed method integrates shortest-path search, dynamic velocity modeling, and time-driven interpolation, thereby establishing a four-stage reconstruction mechanism comprising spatial mapping, path planning, stochastic speed sampling, and temporal interpolation. [Data] Using mobile signaling data from 50 000 users in Shanghai, coupled with roadnetwork topology data, the algorithm successfully generated highly refined trajectories that adhered to road-network constraints. [Conclusion] Compared with the three mainstream methods,(HMM-Reconstruct, Map-Matching Plus and GAT-Traj), the proposed approach reduced the average trajectory error by 33.3%, 21.1%, and 12.5%, respectively, demonstrating superior precision in trajectory reconstruction. The reconstructed trajectories accurately captured spatiotemporal traffic patterns, including rush-hour flow dynamics and polycentric mobility radiation. A cross-city validation(Beijing and Xiamen) confirmed the transferability of this approach, demonstrating its potential for traffic-flow prediction, OD analysis, and smart-city applications.
[1]李文翔，丁龙远，张逸文，等.基于大语言模型的多模式出行方案个性化推荐[J].交通运输工程与信息学报，2025,23(1):93-103.LI Wenxiang, DING Longyuan, ZHANG Yiwen, et al.Personalized recommendation of multimodal travel schemes based on large language models[J]. Journal of Transportation Engineering and Information,2025,23(1):93-103.
[2]NEWSON P, KRUMM J. Hidden Markov map matching through noise and sparseness[C]//Proceedings of the 17th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. Seattle Washington:ACM, 2009:336-343.
[3]KEMPINSKA K, DAVIES T, SHAWE-TAYLOR J. Probabilistic map-matching using particle filters[EB/OL].(2016-11-29)[2025-06-30]. https://doi. org/10.48550/arXiv.1611.097069.
[4]ZHU L, HOLDEN J R, GONDER J D. Trajectory segmentation map-matching approach for large-scale, highresolution GPS data[J]. Transportation Research Record,2017, 2645(1):67-75.
[5]PERRINE K, KHANI A, RUIZ-JURI N. Map-matching algorithm for applications in multimodal transportation network modeling[J]. Transportation Research Record,2015, 2537(1):62-70.
[6]QU L, ZHOU Y, LI J X, et al. HMM-based map matching and spatiotemporal analysis for matching errors with taxi trajectories[J]. ISPRS International Journal of Geo-Information, 2023, 12(8):330.
[7]JIANG L L, CHEN C X, CHEN C. L2MM:learning to map matching with deep models for low-quality GPS trajectory data[J]. ACM Transactions on Knowledge Discovery from Data, 2023, 17(3):1-25.
[8]FENG J, LI Y, ZHAO K, et al. DeepMM:deep learning based map matching with data augmentation[J]. IEEE Transactions on Mobile Computing, 2022, 21(7):2372-2384.
[9]KATARIYA V, BAHARANI M, MORRIS N, et al. DeepTrack:lightweight deep learning for vehicle trajectory prediction in highways[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(10):18927-18936.
[10]QIN X Y, LI Z H, ZHANG K, et al. Vehicle trajectory prediction via urban network modeling[J]. Sensors, 2023,23(10):4893.
[11]YAO Z H, LIU M, JIANG Y S, et al. Trajectory reconstruction for mixed traffic flow with regular, connected,and connected automated vehicles on freeway[J]. IET Intelligent Transport Systems, 2024, 18(3):450-466.
[12]YIN W F, ZHAI J Y, YU Y B. Reconstruction of highway vehicle paths using a two-stage model[J]. Mathematics, 2025, 13(4):618.
[13]ZHANG T T, JIN P J, PICCOLI B, et al. Deep spatialtemporal embedding for vehicle trajectory validation and refinement[J]. Computer-Aided Civil and Infrastructure Engineering, 2024, 39(11):1597-1615.
[14]AYFANTOPOULOU G, MILITSIS M N, GRAU J M S,et al. Improving map matching of floating car data with artificial intelligence techniques[J]. Information, 2022,13(11):508.
[15]ZHANG H Y, WU H, SUN W W, et al. Deeptravel:a neural network based travel time estimation model with auxiliary supervision[EB/OL].(2018-2-6)[2025-06-30].https://doi.org/10.48550/arXiv.1802.02147.
[16]TORTORA M, CORDELLI E, SODA P. PyTrack:a mapmatching-based python toolbox for vehicle trajectory reconstruction[J]. IEEE Access, 2022, 10:112713-112720.
[17]CHAO P F, XU Y H, HUA W, et al. A survey on mapmatching algorithms[C]//Databases Theory and Applications:31st Australasian Database Conference. Melbourne:Springer International Publishing, 2020:121-133.
[18]HUANG Z F, QIAO S J, HAN N, et al. Survey on vehicle map matching techniques[J]. CAAI Transactions on Intelligence Technology, 2021, 6(1):55-71.
[19]HU D L, CHEN L, FANG H X, et al. Spatio-temporal trajectory similarity measures:a comprehensive survey and quantitative study[J]. IEEE Transactions on Knowledge and Data Engineering, 2023, 36(5):2191-2212.
[20]吴红兰，胡德富，郭旭周.基于Informer的车辆多意图运动轨迹预测[J].交通运输工程与信息学报，2024,22(3):68-79.WU Honglan,HU Defu,GUO Xuzhou. Vehicle multi-intention motion trajectory prediction based on informer[J]. Journal of Transportation Engineering and Information,2024,22(3):68-79.
[21]付川云,刘林才,周悦,等.基于GPS轨迹数据的超速事件探测方法试验研究[J].交通运输工程与信息学报,2019,17(2):1-9.FU Chuanyun, LIU Lincai, ZHOU Yue, et al. Experimental research on speeding event detection method based on GPS trajectory data[J]. Journal of Transportation Engineering and Information,2019,17(2):1-9.
[22]吴静娴，杨敏，陈学武，等.基于NestedLogit模型和蒙特卡罗法的通勤者活动链模拟[J].交通运输工程与信息学报, 2016, 14(2):76-82.WU Jingxian, YANG Min, CHEN Xuewu, et al. Simulation of commuter’s activity chain based on nested logit model and monte Carlo method[J].Journal of Transportation Engineering and Information,2016,14(2):76-82.
[23]YAO Y, JUANG Y H, YU Q, et al. A GIS-based framework for synthesizing city-scale long-term individuallevel spatial-temporal mobility[J]. ISPRS International Journal of Geo-Information, 2024, 13(7):261.
[24]LUO A, CHEN S H, XV B. Enhanced map-matching algorithm with a hidden Markov model for mobile phone positioning[J]. ISPRS International Journal of Geo-Information, 2017, 6(11):327.
[25]CHENG H, LIU M M, CHEN L, et al. Gatraj:a graphand attention-based multi-agent trajectory prediction model[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2023, 205:163-175.
基本信息:
DOI：10.19961/j.cnki.1672-4747.2025.06.037
中图分类号:U495
引用信息:
[1]姚尧,蒋应红,邹国建,等.路网约束下基于动态速度模拟的个体轨迹重构[J].交通运输工程与信息学报,2026,24(01):38-49.DOI:10.19961/j.cnki.1672-4747.2025.06.037.
基金信息:
国家重点研发计划项目（2022YFC3801505）;国家重点研发计划项目（2018YFB1601301）; 上海市超级博士后项目（2023045）; 国家自然科学基金项目（71961137006）
2025-08-08
2025-08-08
2025-08-08