Queue length is a crucial measurement of traffic signal control at urban intersections. Conventional queue length estimation methods mostly still rely on fixed detectors. The development of connected vehicles (CV) provides massive amounts of vehicle trajectory data, and the queue length estimation based on CV data has received considerable attention in recent years. However, most existing CV-based methods require the prior knowledge of the penetration rate of CV and vehicle arrivals, but the estimation of these prior distributions has not been well studied. To address this issue, this paper proposes a cycle-based queue length estimation method under partially connected vehicle (CV) environment, with the prior vehicle arrivals being unknown. The empirical Bayes method is adopted to estimate the arrival rate by leveraging the observed queued CV information such as the number and positions. The hyperparameter estimation problem of the prior distribution is solved by the maximum likelihood estimation (MLE) method. To validate the proposed queue length estimation method, a simulation environment with partially connected vehicles is established using VISSIM and Python for data generating. The results in terms of normalized mean absolute errors (NMAE) and normalized root mean square errors (NRMSE) show that the proposed method could produce accurate and reliable estimated queue length under various CV penetration rates.

As a typical spatiotemporal problem, there are three main challenges in traffic forecasting. First, the road network is a nonregular topology, and it is difficult to extract complex spatial dependence accurately. Second, there are short- and long-term dependencies between traffic dates. Third, there are many other factors besides the influence of spatiotemporal dependence, such as semantic characteristics. To address these issues, we propose a spatiotemporal DeepWalk gated recurrent unit model (ST-DWGRU), a deep learning framework that fuses spatial, temporal, and semantic features for traffic speed forecasting. In the framework, the spatial dependency between nodes of an entire road network is extracted by graph convolutional network (GCN), whereas the temporal dependency between speeds is captured by a gated recurrent unit network (GRU). DeepWalk is used to extract semantic information from road networks. Three publicly available datasets with different time granularities of 15, 30, and 60 min are used to validate the short- and long-time prediction effect of this model. The results show that the ST-DWGRU model significantly outperforms the state-of-the-art baselines.

Automatic intersection identification and extraction are an important foundation for urban road network updates and traffic network analysis and modeling. Existing intersection extraction methods based on steering angles and stopping points suffer from inadequate sampling amounts and threshold settings. To address this problem, we propose a road network intersection automatic extraction method based on vehicle trajectory intersection clustering. First, the continuous trajectory segments are extracted from trajectory data based on the sampling interval. Second, the maximum reconstruction error method is developed to extract straight-line trajectory segments from continuous trajectory segments. The overlapped straight-line trajectory segments belonging to the same direction are merged to reduce the number of segments and enhance road network patterns. To further improve the calculation efficiency of the intersection points of straight-line segments, bounding box filtering and orthogonal filtering are used to filter the straight-line trajectory segments that do not have an intersection relationship. Finally, the obtained straight-line segment intersection points are clustered using a density peak clustering algorithm. The road intersections are automatically extracted using the clustering center. The experimental results on real vehicle trajectories in Lianyungang City show that the proposed method performs well on intersection recognition and calculation efficiency.

GPS trajectory data in intersections are series data with different lengths. Dynamic time wrapping (DTW) is good to measure the similarity between series with different lengths, however, traditional DTW could not deal with the inclusive relationship well between series. We propose a unified generalized DTW algorithm (GDTW) by extending the boundary constraint and continuity constraint of DTW and using the weighted local distance to normalize the cumulative distance. Based on the density peak clustering algorithm DPCA using asymmetric GDTW to measure the similarity of two trajectories, we propose an improved DPCA algorithm (ADPC) to adopt this asymmetric similarity measurement. In experiments using the proposed method, the number of clusters is reduced.

Accurate traffic state prediction plays an important role in traffic guidance, travel planning, etc. Due to the existence of complex spatio-temporal relationships, there are some challenges in forecasting. Firstly, in terms of spatial correlation, some models only consider the road network structure information, and ignore the relative location relationships between nodes. Secondly, some models ignore the different impacts of nodes in the global road network on traffic. To solve these problems, we propose a new traffic state-forecasting model, namely, spatio-temporal attention-gated recurrent neural network (ST-AGRNN). In the proposed model, structure-based and location-based localized spatial features are obtained simultaneously by Graph Convolutional Networks (GCNs) and DeepWalk. The localized temporal features are obtained by gated recurrent unit (GRU). The attention-based approach is used to obtain global spatio-temporal features. Experimental validation is performed with two real-world public datasets, and the results show that the ST-AGRNN model outperforms the state-of-the-art methods.

Urban road intersection recognition and feature extraction are crucial for road network modeling and traffic flow analysis. This paper proposes a novel method of intersection recognition and feature extraction based on vehicle trajectory data. First, valid spatial-temporal continuous trajectory segments of each vehicle are obtained by time differences analysis between adjacent trajectory points. Second, based on the basis that the turning behaviors of vehicles mostly occurring at the intersection area, the turning angles are calculated, and the valid turning points are extracted as the samples of clustering algorithm. Finally, the density peak clustering algorithm (DPCA) is used to recognize the intersections from road network which reconstructed from the vehicle trajectory data. The experiment on real-world vehicle navigation trajectories in the city of Lianyungang shows that the proposed method is able to recognize intersection accurately with scalability.