The widespread deployment of road sensors in the Internet of Things (IoT) allows for fine-grained data integration, which is a fundamental demand for data-driven applications. Sensing data with inevitable missing and substantial anomalies are unavoidable, due to unstable network communication, faulty sensors, etc. Recent tensor completion studies have demonstrated the superiority of deep learning in imputation tasks by precisely capturing the intricate spatiotemporal dependencies/correlations. However, ignoring the significance of initial interpolation in these methods results in unstable performance, especially for complicated missing scenarios across large-scale data. Additionally, the existing interpolation methods utilize recursive signal propagation along spatiotemporal dimensions, which produce noise accumulation where the dependencies are uncorrelated. In this study, we design a multiattention tensor completion network (MATCN) for modeling multidimensional representation in the presence of missing entries. MATCN sparsely sampled historical fragments and utilized a gated diffusion convolution layer to generate the initial schemes, which mitigate the exposure bias existing in previous traffic imputation models. In addition, we develop a spatial signal propagation module and a temporal self-attention module as the basic stack block of deep networks, which executes representation aggregation and dynamic dependencies extraction at the spatiotemporal level. This architecture empowers MATCN with progressive completion capacities for complex data missing scenarios. Numerical experiments on four real-world traffic data sets with various missing scenarios demonstrate the superiority of MATCN over multiple state-of-the-art imputation baselines.
Global positioning system (GPS) trajectory map matching projects GPS coordinates to the road network. Most existing algorithms focus on the geometric and topological relationships of the road network, while did not make full use of the historical road network information and floating car data. In this study, the authors proposed a deep learning enabled vehicle trajectory map‐matching method with advanced spatial–temporal analysis (DST‐MM). The algorithm mainly focused on the following three aspects: (i) analyse the spatial relevancy from the prospective of geometric analysis, topology analysis and intersection analysis; (ii) to make full use of the historical and real‐time data, a deep learning model was conducted to extract the road network and vehicle trajectory features and (iii) establish a speed prediction model and nest it in the temporal analysis structure. It narrows down the path search range through establishing the dynamic candidate graph. Experimental results show that the proposed DST‐MM algorithm outperforms the existing algorithms in terms of matching accuracy for low‐sampling frequencies GPS data, especially in the central urban area.
16 February 2025 in Expert Systems with Applications