Low-Rank Hankel Tensor Completion for Traffic Speed Estimation

被引:12
|
作者
Wang, Xudong [1 ]
Wu, Yuankai [1 ]
Zhuang, Dingyi [1 ]
Sun, Lijun [1 ]
机构
[1] McGill Univ, Dept Civil Engn, Montreal, PQ H3A 0C3, Canada
基金
加拿大创新基金会; 加拿大自然科学与工程研究理事会;
关键词
Spatio-temporal traffic data; traffic state esti-mation; missing data imputation; low-rank tensor completion; delay embedding transform; STATE ESTIMATION; MATRIX COMPLETION; FILTER; FLOW;
D O I
10.1109/TITS.2023.3247961
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
This paper studies the traffic state estimation (TSE) problem using sparse observations from mobile sensors. Most existing TSE methods either rely on well-defined physical traffic flow models or require large amounts of simulation data as input to train machine learning models. Different from previous studies, we propose a purely data-driven and model-free solution in this paper. We consider the TSE as a spatiotemporal matrix completion/interpolation problem, and apply spatiotemporal delay embedding to transform the original incomplete matrix into a fourth-order Hankel structured tensor. By imposing a low-rank assumption on this tensor structure, we can approximate and characterize both global and local spatiotemporal patterns in a data-driven manner. We use the truncated nuclear norm of a balanced spatiotemporal unfolding -- in which each column represents the vectorization of a small patch in the original matrix -- to approximate the tensor rank. An efficient solution algorithm based on the Alternating Direction Method of Multipliers (ADMM) is developed for model learning. The proposed framework only involves two hyperparameters, spatial and temporal window lengths, which are easy to set given the degree of data sparsity. We conduct numerical experiments on real-world high-resolution trajectory data, and our results demonstrate the effectiveness and superiority of the proposed model in some challenging scenarios.
引用
收藏
页码:4862 / 4871
页数:10
相关论文
共 50 条
  • [21] PARALLEL MATRIX FACTORIZATION FOR LOW-RANK TENSOR COMPLETION
    Xu, Yangyang
    Hao, Ruru
    Yin, Wotao
    Su, Zhixun
    [J]. INVERSE PROBLEMS AND IMAGING, 2015, 9 (02) : 601 - 624
  • [22] Union of low-rank tensor spaces: Clustering and completion
    Ashraphijuo, Morteza
    Wang, Xiaodong
    [J]. Journal of Machine Learning Research, 2020, 21
  • [23] PROVABLE MODELS FOR ROBUST LOW-RANK TENSOR COMPLETION
    Huang, Bo
    Mu, Cun
    Goldfarb, Donald
    Wright, John
    [J]. PACIFIC JOURNAL OF OPTIMIZATION, 2015, 11 (02): : 339 - 364
  • [24] Tensor Completion via Nonlocal Low-Rank Regularization
    Xie, Ting
    Li, Shutao
    Fang, Leyuan
    Liu, Licheng
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2019, 49 (06) : 2344 - 2354
  • [26] Union of Low-Rank Tensor Spaces: Clustering and Completion
    Ashraphijuo, Morteza
    Wang, Xiaodong
    [J]. JOURNAL OF MACHINE LEARNING RESEARCH, 2020, 21
  • [27] Low-Rank tensor completion based on nonconvex regularization
    Su, Xinhua
    Ge, Huanmin
    Liu, Zeting
    Shen, Yanfei
    [J]. SIGNAL PROCESSING, 2023, 212
  • [28] Nonlocal Low-Rank Tensor Completion for Visual Data
    Zhang, Lefei
    Song, Liangchen
    Du, Bo
    Zhang, Yipeng
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2021, 51 (02) : 673 - 685
  • [29] Low-rank Tensor Completion for PMU Data Recovery
    Ghasemkhani, Amir
    Liu, Yunchuan
    Yang, Lei
    [J]. 2021 IEEE POWER & ENERGY SOCIETY INNOVATIVE SMART GRID TECHNOLOGIES CONFERENCE (ISGT), 2021,
  • [30] Robust approximations of low-rank minimization for tensor completion
    Gao, Shangqi
    Zhuang, Xiahai
    [J]. NEUROCOMPUTING, 2020, 379 : 319 - 333