Urban Expressway Research Articles

A dynamic speed guidance method at on-ramp merging areas of urban expressway considering driving styles

A dynamic speed guidance method at on-ramp merging areas of urban expressway considering driving styles

A traffic state prediction method based on spatial–temporal data mining of floating car data by using autoformer architecture

AbstractFloating car data (FCD), characterized by wide spatiotemporal coverage, low collection cost, and immunity to adverse weather conditions, are one of the key approaches for intelligent transportation systems to obtain real‐time urban road network traffic information. The research aims to utilize GPS data from taxis in Shanghai and vector geographic information data of the road network, with urban expressways as the research focus. Based on the different driving characteristics of expressways and the vehicles on the ramps below, a clustering analysis is employed to determine all floating vehicles traveling on the target road. Furthermore, an adaptive buffer zone consistent with the road orientation is established based on road vector geographic data. This allows for the extraction of FCD within segmented areas, and the average vehicle speed for that road segment is obtained through weighted calculations. This method fully exploits the natural characteristics of taxis in urban areas with a wide spatiotemporal distribution. The data effectiveness and coverage reach 90.2% and 85.7%, respectively, significantly surpassing the traditional grid‐based extraction method for FCD. Additionally, to capture the long‐term spatiotemporal dependencies of road network traffic states, a spatial–temporal autoformer (STAF) network based on spatial–temporal sequence autocorrelation is employed for traffic state prediction. The results indicate that the STAF method demonstrates good performance in medium‐ and long‐term prediction. We believe that the proposed FCD mining method in this paper provides a new approach for efficiently extracting large‐scale road network traffic states and conducting medium‐ to long‐term predictions.

Freeway congestion management on multiple consecutive bottlenecks with RL‐based headway control of autonomous vehicles

AbstractAdaptive cruise control (ACC) is the core building block of future full autonomous driving. Numerous recent research demonstrated that Autonomous Vehicles (AVs) adopting shorter headways generally increase road capacity and may relieve congestion at bottlenecks for moderate demand scenarios. However, with high demand scenarios, bottlenecks can still be activated causing capacity breakdown. Therefore, extra control measures as dynamic traffic control near bottlenecks is necessary. The challenge is harder on urban freeways with consecutive bottlenecks which affect each other. This paper aims to improve the performance of ACC systems in a high demand scenario. A multi‐bottleneck dynamic headway control strategy based on deep reinforcement learning (DRL) that adapts headways to optimize traffic flow and minimize delay is proposed. The controller dynamically assigns an optimal headway for each controlled section, based on state measurement representing the current traffic conditions. The case study is a freeway stretch with three consecutive bottlenecks which is then extended to include eight bottlenecks. Three different RL agent configurations are presented and compared. It is quantitatively demonstrated that the proposed control strategy improves traffic and enhances the system delay by up to 22.30%, and 18.87% compared to shortest headway setting for the three‐bottleneck and the eight‐bottleneck networks, respectively.

Impact of different grade roads on ecological networks: A case study of Fuzhou City, China.

The expansion of roads exacerbates the fragmentation of ecological networks and obstructs landscape connectivity. Scientific analysis of the impacts of different grades of roads on landscape connectivity and ecological networks is crucial for guiding road planning and ecological conservation. Based on the data of 2020 road network, land cover types, and digital elevation models, we used morphological spatial pattern analysis and circuit theory to construct ecological networks within different species dispersal distances (1, 3, 5, 10 km) in Fuzhou. We analyzed the impacts of roads of different grades (motorway, urban expressway, primary and secondary highway) on landscape connectivity at the landscape-patch-corridor scale. The results showed that at the landscape scale, overall landscape connectivity was significantly positively correlated with species dispersal distance. The motorway, urban expressway, primary and secondary highway had the lowest decline rate of overall landscape connectivity within a 10 km species dispersal range, being reduced by 15.6%, 5.3%, 1.5% and 5.2%, respectively. At the patch scale, in the comparison of roads of different grades, motorway led to the highest decline rate of patch connectivity within 1 and 5 km species dispersal range, while primary highway led to the highest decline rate of patch connectivity within 3 and 10 km species dispersal range. At the corridor scale, urban expressway led the highest increase rate of indices. The cost-weighted distance of the overall least-cost path, the ratio of cost-weighted distance to length, ove-rall effective resistance, and total corridor length within 5 km species dispersal range were increased by 43.4%, 33.2%, 57.3%, and 7.3%, respectively. As the distance of species dispersal increased, the patches with high importance were reduced from the northern, central, and northwestern regions to the northern regions, leading to a decrease in the living space of species, and the key corridors were gradually extending from the northwestern and southern regions to the central regions. Our results can guide the construction and optimization of Fuzhou's ecological network from an overall perspective, and provide a scientific basis for biodiversity conservation, ecological restoration, and road network planning under the context of limited land resource utilization.

Coordinated Control of Urban Expressway Integrating Adjacent Signalized Intersections Using Adversarial Network Based Reinforcement Learning Method

Coordinated Control of Urban Expressway Integrating Adjacent Signalized Intersections Using Adversarial Network Based Reinforcement Learning Method

Prediction of driving stress on high-altitude expressway using driving environment features: A naturalistic driving study in Tibet

Objective Owing to the harsh environment in high-altitude areas, drivers experience significant driving stress. Compared with urban roads or expressways in low-altitude areas, the driving environment in high-altitude areas has distinct features, including mountainous environments and a higher proportion of trucks and buses. This study aims to investigate the feasibility of predicting stress levels through elements in the driving environment. Methods Naturalistic driving tests were conducted on an expressway in Tibet. Driving stress was assessed using heart rate variability (HRV)-based indicators and classified using K-means clustering. A DeepLabv3 model was built to conduct semantic segmentation and extract environment elements from the driving scenarios recorded through a camera next to the driver’s eyes. A decision tree and 4 other ensemble learning models based on decision trees were built to predict driving stress levels using the environment elements. Results Fifty-six indicators were extracted from the driving environment. Results of the prediction models demonstrate that extreme gradient boosting has the best overall performance with the F1 score (harmonic mean of the precision and recall) and G-mean (geometric mean of sensitivity and specificity) reaching 0.855 and 0.890, respectively. Indicators based on the variation rate of trucks and buses have high feature importance and exhibit positive effects on driving stress. Indicators reflecting the proportion of mountain, road, and sky features negatively affect the expected levels of driving stress. Additionally, the mountain feature demonstrates multidimensional effects, because driving stress is positively affected by indicators of the variation rate for mountain elements. Conclusions This study validates the prediction of driving stress using environment elements in the driver’s field of view and extends its application to high-altitude expressways with distinct environmental characteristics. This method provides a real-time, less intrusive, and safer method of driving stress assessment and prediction and also enhances the understanding of the environmental determinants of driving stress. The results hold promising applications, including the development of a driving state assessment and warning module as well as the identification of high-risk road sections and implementation of control measures.

Reliability Analysis of Horizontal Curves Using Geometric Design Consistency Assessment Criterion

Road accidents have always been one of the important reasons for fatalities and financial losses. Since road accidents on rural highways cause more serious injuries than those on urban highways, providing a suitable method to increase safety in the curves can be a significant contributor to preventing these damages. Although speed is one of the most important variables affecting highway safety, numerous studies have been performed on the reliability analysis of horizontal curves without taking the speed variable into account. The aim of this research is reliability (probability of noncompliance) assessment in the horizontal curve design using geometric design consistency criteria. The radius, superelevation, and operating speed of 19 horizontal curves were collected by field research on the Mashhad‐Torbat Heydarieh highway in Iran. Three different approaches were defined based on the geometric design consistency criterion of a single horizontal curve, and consecutively, the probability of noncompliance was calculated using these approaches. According to the obtained results, this study showed that radius enhancement increases the probability of noncompliance and the consistency level of the geometric design. Finally, the high values of the probability of noncompliance (failure) indicate that the geometric design guidelines need calibration in the design of horizontal curves, especially for higher radii.

Energy Efficiency Analysis and Research of Pure Electric Vehicles under Actual Working Conditions

Pure electric vehicles are a hot topic in today’s social development. From beginning to end, the energy density of power batteries in pure electric vehicles is generally low, resulting in short range, inaccurate range estimation, and driver anxiety. This is also one of the reasons that hinder the promotion of pure electric vehicles. This article discusses the energy consumption of pure electric vehicles based on actual road tests. By comparing the battery energy efficiency of various roads (including urban roads, suburban roads, and highways) using electric vehicles from Guangzhou to Qingyuan, the energy consumption patterns of pure electric vehicles are analyzed. Experiments have shown that continuously improving the energy density of power batteries is the key technology to solve the problem of insufficient range for electric vehicles. Based on the existing power battery technology, this study investigates the relationship between energy consumption and various factors affecting electric vehicles, which can provide better reference for drivers to predict energy consumption and range.

Investigating the effect of dynamic traffic distribution on network-wide traffic emissions: An empirical study in Ningbo, China.

Urban road traffic is one of the primary sources of carbon emissions. Previous studies have demonstrated the close relationship between traffic flow characteristics and carbon emissions (CO2). However, the impact of dynamic traffic distribution on carbon emissions is rarely empirically studied on the network level. To fill this gap, this study proposes a dynamic network carbon emissions estimation method. The network-level traffic emissions are estimated by combining macroscopic emission models and recent advances in dynamic network traffic flow modeling, namely, Macroscopic Fundamental Diagram. The impact of traffic distribution and the penetration of battery electric vehicles on total network emissions are further investigated using the Monte Carlo method. The results indicate the substantial effect of network traffic distribution on carbon emissions. Using the urban expressway network in Ningbo as an example, in the scenario of 100% internal combustion engine vehicles, increasing the standard deviation of link-level traffic density from 0 to 15 veh/km-ln can result in an 8.9% network capacity drop and a 15.5% reduction in network carbon emissions. This effect can be moderated as the penetration rate of battery electric vehicles increases. Based on the empirical and simulating evidence, different expressway pollution management strategies can be implemented, such as petrol vehicle restrictions, ramp metering, congestion pricing, and perimeter control strategies.

Hybrid System Analysis for Surface Rainwater Drainage in Urban Highway Underpasses: The Case of Diyarbakir City

Dünyada meydana gelen birçok trafik kazalarındaki “boğulma ölümleri” genellikle karayollarının eğimli alt geçitlerinde veya çöküntülü bölümlerinde meydana gelmektedir. Bir kentsel alanda yer alan karayolu alt geçit noktalarından yağmur suyunun hızlı ve güvenli bir şekilde nasıl uzaklaştırılacağı konusunda artan bir endişe bulunmakla birlikte bu sorunun çözümleri de aranmaktadır. Bu çalışmada Diyarbakır- Şanlıurfa kent içi karayolu üzerinde bulunan altgeçit-köprülü kavşaklardan Doğum Hastanesi Köprülü Kavşağı uygulama alanı olarak seçilmiş olup hem güvenlik hem de hidrolik verimlilik açısından karayolu alt geçidi drenaj sistemini iyileştirmek için yeni bir hibrit sistem tasarımı araştırılmıştır. Bu çerçevede alt geçit drenaj tesisinin iki sistemden oluşması tavsiye edilmektedir. Bu ikili sistem; küçük fırtına olayını yerçekimi yoluyla toplamak için boyutlandırılmış bir yeraltı kanalizasyon hattı ve büyük bir fırtına olayı sırasında fazla suyu drene edecek bir pompa sisteminden oluşmaktadır. Pompa sistemi, acil durumlarda devreye girecek sabit bir istasyon veya hareketli bir birimden oluşmaktadır. Birden fazla fırtına olayı için böyle bir hibrit drenaj sisteminin tasarımında, kanalizasyon borularını ve karter pompalarını boyutlandırmak için sırasıyla kullanılan Rasyonel Metot ve önerilen sistemin performansını iki fırtına olayı altında doğrulamak için benimsenen EPA SWMM bilgisayar modeli olmak üzere iki hidrolojik yaklaşım kullanılmaktadır. Bu araştırmada seçilen alt geçit noktalarındaki hidrolik verimlilik analizleri için bu metotlarla tasarım senaryoları çalışılmıştır. Sonuçlar irdelendiğinde mevcut yağmur suyu drenaj sisteminin olası 5-100 yıllık tekerrür eden yağış olaylarında yetersiz kaldığı görülmüştür. Böylece yapılan simülasyon neticesinde mevcut yağmur suyu drenaj sisteminin kapasitesi artırılmış ve sisteme pompa entegrasyonu sağlanmasıyla yağmur suyu drenaj sistemi kapasitesinin yeterli bir seviyeye geldiği gözlemlenmiştir.

Assessment of ALINEA method performance at different loop detector locations using field data and micro-simulation modeling via AIMSUN

Abstract A ramp metering (RM) or ramp signal is an effective method for traffic management, especially at merging sections in urban expressways. This research aims to assess the Asservissement Linéaire d’Entrée Autoroutière (ALINEA) method performance at different loop detector locations using field data and micro-simulation modeling via advanced interactive microscopic simulator for urban and non-urban networks (AIMSUN). The operation of RM was executed using the ALINEA method. The field data were collected from two locations in Manchester in the UK for calibration and validation in AIMSUN. Occupancy values at different loop detector locations were evaluated. The results indicated that the current simulation model could represent reality to acceptable limits. Increasing the number of detector locations at the merging entrance will decrease the occupancy value. The results showed that the effect of loop detector location is a significant factor in determining the critical occupancy of the merging section. Also, the downstream location is more appropriate for determining the congestion or critical status. The AIMSUN software showed significant consistency in representing field data of merging sections. The minimum value of GEH (Geoffrey E. Havers) that equal to 0.5 has been used to statistically and graphically demonstrate this behavior.

Development laws of geological hazards along urban highway in Southwest China and countermeasures for prevention and control

China is a mountainous country with highly developed road geologic hazards, which pose a great threat to the construction and operation of highways, bridges, and tunnels and to the safety of people and property. This paper discussed the types, basic features, formation, and prevention conditions of road geologic hazards in China based on field research and study data collected thus far. The study considered an urban area of a city in southwest China as the center and a geological field investigation was performed over a total of 282 ​km on three important lifeline projects. The results show: Types of geologic hazards along the highways are mainly avalanches, debris flows, and landslides, respectively. Among them, the landslips are mainly distributed along the roads, with slip, dumping, and wrong break types as the main ones; the debris flows are widely distributed, mainly concentrated in the river valleys; and the unstable slopes are relatively few in number. Geological disasters are characterized by large-scale and concentrated triggering in time and space, and a single disaster can easily trigger other disasters, thus forming a chain of disasters. Neotectonic movement, seismic activity, topography, climatic conditions, stratigraphic lithology, and human activities are important factors leading to geologic hazards in the study area. This study is of great practical significance for reducing the occurrence of roadbed diseases and prolonging the service life of highways.

Effects of Exclusive Lanes for Autonomous Vehicles on Urban Expressways under Mixed Traffic of Autonomous and Human-Driven Vehicles

Autonomous vehicles (AVs) are expected to be seen easily on our roads, providing benefits to society in the future. However, due to the challenges faced such as technical and regulatory issues, it will take a long time to reach 100% AVs, suggesting the coexistence of AVs and human-driven vehicles (HVs) for a substantial time. This coexistence of the mixed traffic of AVs and HVs may affect network performances in urban areas. Under this circumstance, to promote the rapid advancement of AV technology, transportation authorities are considering introducing AV-exclusive lanes. This study evaluates AV lane operation scenarios with mixed traffic consisting of both AVs and HVs. We utilized a macroscopic traffic simulation-based approach to measure the effects of AV lanes on two urban expressways in Seoul, South Korea. We varied the market penetration rate (MPR) from 10 to 80% in 10% increments. The simulation results indicate that the impact of exclusive lanes can be either negligible or beneficial depending on the characteristics of the roads. The results also showed that, in terms of the efficient use of road capacity, AV lanes should be introduced when the MPR reaches at least 20 to 30%. The results also showed that the capacity gain effect is more sensitive over a lower MPR range. We expect this study to not only enhance policy makers’ and planners’ fundamental understanding of the early stages of AV-exclusive lanes, but also help them prepare a large-scale urban network structure in the future.

Interchange and Comact Grade Separated Junction: Classifications' System Considerations, Traffic Safety, and Operations Overview

The paper presents system considerations and classification criteria of interchange geometric design based on the new Israeli Interchange design guidelines (VOLUME III of geometric design standards for rural (interurban) highways and urban freeways).
 The paper gives a methodological overview of interchanges systematical warrant in order to adjust and essentially construct an appropriate interchange that connects two intersecting highways. This warrant refers to the number of legs, partiality, functional classification, and access control type of intersecting highways zone (in the final stage of construction) based on highway classification. The paper also covers system considerations for implementing a compact grade separated junction (CGSJ) in a major highway, and traffic operation, safety, and public transport insights.

Bayesian-optimized interpretable surrogate model for seismic demand prediction of urban highway bridges

Urban highway bridges are crucial for regional economic development and population mobility, but they are vulnerable to seismic hazards. Accurately predicting the seismic demands of bridges is essential. Several commonly-used methods for seismic demand estimates, such as cloud analysis, incremental dynamic analysis, and multiple stripe analysis, are computationally expensive and subject to theoretical assumptions and convergence issues, especially for portfolios of bridges. Machine learning techniques can be effective, but their lack of interpretability hinders understanding of the surrogate model's predictions. To address these challenges, this study develops a Bayesian-optimized interpretable ensemble learning surrogate model for predicting and interpreting the critical seismic demands of urban highway bridges.Thousands of continuous prestressed concrete girder bridges with typical geometric configurations in China are generated, and 120 ground motion records with different frequency, waveform, and duration characteristics are chosen. The seismic demand dataset is established based on nonlinear time-history analyses for urban highway bridges, accounting for different types of uncertainties. The surrogate models are developed based on the XGBoost ensemble learning method, with hyperparameters automatically and efficiently determined by Bayesian optimization technique. Compared to the random search and grid search methods, Bayesian optimization is approximately 8 times and 29 times faster, respectively, and it also achieves the highest prediction accuracy. Meanwhile, the relationship of each hyperparameter in the surrogate model is also revealed. The prediction results of the proposed model are compared with other machine learning models. The SHAP algorithm is subsequently used to explain the model predictions regarding the feature importance, global interpretations, and feature dependency analysis. The study quantifies the contribution of each structural feature and seismic intensity measure to the selected critical seismic demand and identifies the interaction effects between different input variables in the prediction.

The impact of lockdown, fatigue, and social interaction on highway demand during the COVID-19 pandemic: The case of Santiago, Chile

The COVID-19 pandemic, as well as the government measures to curb its spread, have significantly affected mobility. Various studies have investigated behavioral changes across different transport modes accounting for sociodemographic variables, yet the focus has predominantly been on public transportation. This article addresses this gap by quantifying the impact of mobility restrictions on an urban highway in Santiago, Chile. To do so, we develop several econometric models based on panel data, which enable us to assess control measures’ effects while accounting for their spatial heterogeneity. Our computational experiments demonstrate that traffic reductions were more significant among higher-income drivers (1% reduction per each 100,000 pesos increase). Conversely, municipalities with a higher proportion of elderly residents saw less drastic decreases in traffic. Regarding the effectiveness of control measures, we confirm that the lockdown is affected by fatigue and social interaction. The fatigue implies that users do not fulfill the lockdown as time passes, reducing the effect of quarantines by about 50% during October, 2020. The social interaction effect suggests the lockdown is less effective when not all city’s municipalities are restricted to travel, which might be an argument against dynamic or partial lockdowns.

Smart and efficient EV charging navigation scheme in vehicular edge computing networks

With the increasing number of electric fast charging stations (FCSs) deployed along roadsides of both urban roads and highways, the long-distance travel of electric vehicles (EVs) becomes possible. The EV charging navigation scheme is significant for the quality of user experience. However, the variable conditions of both power grid and traffic networks make it a serious challenge. In this paper, we propose an efficient EV charging navigation scheme while considering both the electric and computation resource sharing. With the support of vehicular edge computing networks in intelligent transportation systems (ITSs), EVs perform both the flexible power load and edge computing nodes. When the traffic network in the established route starts to become congested, EVs can select to enter the nearest FCS. In addition to being supplemented by electric resources, EVs also benefit by sharing their own computing resource with FCSs. We formulate the EV charging navigation as a mixed integer programming problem, the EV moving route planning, FCS selection, and staying time in FCSs are optimized, to balance the relationships among the traveling time, traveling cost and reward. To address the influence caused by the randomness of traffic conditions and charging prices, a two-stage charging navigation algorithm combined with A∗\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$A^{*}$$\\end{document} algorithm and deep reinforcement learning (DRL) is proposed, with a novel designed reward function. Eventually, numerous experimental results show the effectiveness of the proposed schemes.

Network-wide safety impacts of dedicated lanes for connected and autonomous vehicles

Cooperative, Connected and Automated Mobility (CCAM) enabled by Connected and Autonomous Vehicles (CAVs) has potential to change future transport systems. The findings from previous studies suggest that these technologies will improve traffic flow, reduce travel time and delays. Furthermore, these CAVs will be safer compared to existing vehicles. As these vehicles may have the ability to travel at a higher speed and with shorter headways, it has been argued that infrastructure-based measures are required to optimise traffic flow and road user comfort. One of these measures is the use of a dedicated lane for CAVs on urban highways and arterials and constitutes the focus of this research. As the potential impact on safety is unclear, the present study aims to evaluate the safety impacts of dedicated lanes for CAVs. A calibrated and validated microsimulation model developed in AIMSUN was used to simulate and produce safety results. These results were analysed with the help of the Surrogate Safety Assessment Model (SSAM). The model includes human-driven vehicles (HDVs), 1st generation and 2nd generation autonomous vehicles (AVs) with different sets of parameters leading to different movement behaviour. The model uses a variety of cases in which a dedicated lane is provided at different type of lanes (inner and outer) of highways to understand the safety effects. The model also tries to understand the minimum required market penetration rate (MPR) of CAVs for a better movement of traffic on dedicated lanes. It was observed in the models that although at low penetration rates of CAVs (around 20%) dedicated lanes might not be advantageous, a reduction of 53% to 58% in traffic conflicts is achieved with the introduction of dedicated lanes in high CAV MPRs. In addition, traffic crashes estimated from traffic conflicts are reduced up to 48% with the CAVs. The simulation results revealed that with dedicated lane, the combination of 40-40-20 (i.e., 40% human-driven – 40% 1st generation AVs– 20% 2nd generation AVs) could be the optimum MPR for CAVs to achieve the best safety benefits. The findings in this study provide useful insight into the safety impacts of dedicated lanes for CAVs and could be used to develop a policy support tool for local authorities and practitioners.

Analysis of Traffic Volume and Travel-Time Relationship Using Continuous One-Hour Values on Urban Expressway

To make more efficient use of the expanded freeway and urban expressway networks, various measures such as bottleneck management and wide-area congestion pricing based on traffic data obtained from traffic detectors, including traffic volume and travel time, have been considered. Generally, the congestion status of the data varies from day to day. This study proposes a method for analyzing a graph of traffic volume and travel time to visually and intuitively grasp the change in the daily traffic situation using continuous one-hour values. These values are continuously generated hourly values obtained by shifting data every minute. Twenty-four hours 1 minute data for 128 days on 32 segments with detectors in the Nagoya Expressway Network in Japan were used to draw a continuous one-hour value graph. A number of graphs showed loops of continuous one-hour values with congestion and a smooth variation characteristic of values over time. These graphs provide an accurate estimate of the daily maximum one-hour traffic volumes and facilitate a sequential understanding of the congestion pattern changes on successive route segments. Hourly travel-time prediction models were constructed to macroscopically examine congestion measures over a range of several hours. These models were fabricated with high accuracy using multiple regression analysis based on the characteristics of continuous one-hour values. Exploratory predictive analysis of hourly travel-time models has allowed us to study and discuss various congestion factors in road structures and traffic flows, and it has been found to be easy to grasp the phenomenon and ensure accuracy and operability.

Lane‐level short‐term travel speed prediction for urban expressways: An attentive spatio‐temporal deep learning approach

AbstractNumerous efforts have been made to address the section‐level travel speed prediction problem. However, section‐level predictions can hardly be used for fine‐grained applications, such as lane management and lane‐level navigation. The main reason for this is that significant speed heterogeneity exists among the lanes within one section. Thus, this study proposes a three‐dimensional (3D) dual attention convolution‐based deep learning model for predicting the lane‐level travel speed. 3D convolutions are designed to learn high‐dimensional spatiotemporal traffic flow features, that is, the relationships between different sections, lanes, and periods. Dual attention modules are used to focus on the traffic flow propagation patterns and to explain the model's mechanisms. To evaluate the proposed model, an indicator is introduced to assess the spatio‐temporal learning ability, based on targeting the lane‐level case. Evaluation experiments are conducted based on loop detector data in Shanghai, China. The results show that high accuracy is obtained by the proposed model, with a 2.9 km/h mean absolute error, thereby outperforming several existing methods. Finally, an in‐depth analysis is provided regarding the attention coefficients and interpretation of real‐world lane‐level traffic flow propagation patterns, so as to gain insights into the model's mechanism when capturing dynamic lane‐level traffic flow.