Search

A major shortcoming of the conventional car-following models is that these models only consider the current spacing and speeds of the target vehicle and its immediate leading vehicle, without taking into account prior driving actions, even for those from the same driver. In other words, the numerous prior experiences have no influence in predicting vehicular movements for the next time step. In this research, we propose a machine-learning-based data-driven methodology that is able to take advantage of the high-resolution historical traffic data in the current data-rich era, to predict vehicular movements in an accurate manner with high computational efficiency. The proposed car-following model has a simple model structure based on a fixed-radius near neighbors (FRNN) search algorithm and it can be applied to high-resolution, real-time vehicle movement prediction, modeling, and control. A comprehensive performance comparison is also conducted among the proposed car-following model, another similar data-driven model, and two conventional formula-based models. The results indicate that the FRNN algorithm-based car-following model is superior to all other three models in terms of prediction accuracy and is more computationally efficient compared to its data-driven-based counterpart. Some extensive applications of the proposed car-following model are also discussed at the end of this article.

In the area of urban transportation networks, a growing number of day-to-day (DTD) traffic dynamic theories have been proposed to describe the network flow evolution, and an increasing amount of laboratory experiments have been conducted to observe travelers' behavior regularities. However, the "communication" between theorists and experimentalists has not been made well. This paper devotes to 1) detecting unanticipated behavior regularities by conducting a series of laboratory experiments, and 2) improving existing DTD dynamics theories by embedding the observed behavior regularities into a route choice model. First, 312 subjects participated in one of the eight decision-making scenarios and make route choices repeatedly in congestible parallel-route networks. Second, three route-switching behavior patterns that cannot be fully explained by the classic route-choice models are observed. Third, to enrich the explanation power of a discrete route-choice model, behavioral assumptions of route-dependent attractions, i.e., route-dependent inertia and preference, are introduced. An analytical DTD dynamic model is accordingly proposed and proven to steadily converge to a unique equilibrium state. Finally, the proposed DTD model could satisfactorily reproduce the observations in various datasets. The research results can help transportation science theorists to make the best use of laboratory experimentation and to build network equilibrium or DTD dynamic models with both real behavioral basis and neat mathematical properties.

With automated driving forthcoming, lane-changing for Connected and Automated Vehicles (CAVs) has received wide attention. The main challenge is that lane-changing requires not only local CAV control but also interactions with the surrounding traffic. Nevertheless, the Line-of-Sight (LoS) sensing range of the CAVs imposes severe limitations on lane-changing safety, and the lane-changing decision that is made based only on self-interest ignores its impact on the traffic flow efficiency. To overcome these difficulties, this article proposes a Digital Twin (DT) empowered mobile edge computing (MEC) architecture. With MEC, the sensing and computing capabilities of the CAVs can be strengthened to guarantee real-time safety. The virtualization and offline learning capabilities of the DT can be leveraged to enable the CAVs to learn from the experience of the physical MEC network and make lane-changing decisions via a 'foresight intelligent' approach. A case study of lane-changing is provided where the DT is constituted by a cellular automata based road traffic simulator coupled with a LTE-V based MEC network simulator. Deep reinforcement learning is adopted to train the lane-changing strategy and results validate the effectiveness of our proposed architecture.

The ever-increasing and unbalanced traffic load in cellular vehicle-to-everything (C-V2X) networks have increased the network congestion and led to user dissatisfaction. To relieve the network congestion and improve the traffic load balance, in this paper, we propose an intelligent software defined C-V2X network framework to enable flexible and low-complexity traffic offloading by decoupling the network data plane from the control plane. In the data plane, the cellular traffic offloading and the vehicle assisted traffic offloading are jointly performed. In the control plane, deep learning is deployed to reduce the software defined network (SDN) control complexity and improve the traffic offloading efficiency. Under the proposed framework, we investigate the traffic offloading problem, which can be formulated as a multi-objective optimization problem. Specifically, the first objective maximizes the cellular access point (AP) throughput with consideration of the load balance by associating the users with the APs. The second objective maximizes the vehicle throughput with consideration of the vehicle trajectory by associating the delay-insensitive users with the vehicles. The two objectives are coupled by the association between the cellular APs and the vehicles. A deep learning based online-offline approach is proposed to solve the multi-objective optimization problem. The online stage decouples the optimization problem into two sub-problems and utilizes the ‘Pareto optimal’ to find the solutions. The offline stage utilizes deep learning to learn from the historical optimization information of the online stage and helps predict the optimal solutions with reduced complexity. Numerical results are provided to validate the advantages of our proposed traffic offloading approach via deep learning in C-V2X networks.

The increasing population in cities induces a high travel demand. Unfortunately, due to the limited capacity of urban transport networks, this increasing demand for travel raises various problems a...

Conventional reinforcement learning (RL) models of variable speed limit (VSL) control systems (and traffic control systems in general) cannot be trained in real traffic process because new control actions are usually explored randomly, which may result in high costs (delays) due to exploration and learning. For this reason, existing RL-based VSL control approaches need a traffic simulator for training. However, the performance of those approaches are dependent on the accuracy of the simulators. This paper proposes a new RL-based VSL control approach to overcome the aforementioned problems. The proposed VSL control approach is designed to improve traffic efficiency by using VSLs against freeway jam waves. It applies an iterative training framework, where the optimal control policy is updated by exploring new control actions both online and offline in each iteration. The explored control actions are evaluated in real traffic process, thus it avoids that the RL model learns only from a traffic simulator. The proposed VSL control approach is tested using a macroscopic traffic simulation model to represent real world traffic flow dynamics. By comparing with existing VSL control approaches, the proposed approach is demonstrated to have advantages in the following two aspects: (i) it alleviates the impact of model mismatch, which occurs in both model-based VSL control approaches and existing RL-based VSL control approaches, via replacing knowledge from the models by knowledge from the real process, and (ii) it significantly reduces the exploration and learning costs compared to existing RL-based VSL control approaches.

In the era of information and communications technology and big data, roadway traffic is being monitored by various sensors. Notably, sensors equipped in vehicles, as shown in Figure 1, are providi...

An input vector composed of various features plays an important role in short-term traffic forecasting. However, there is limited research on the optimal feature selection of an input vector for a certain forecasting task. To fill the gap, this paper proposes a cohesion-based heuristic feature selection method by analyzing the nature of the forecasting methods. This method is able to determine which features should be contained in an input vector to make a forecasting algorithm perform better. The proposed method is demonstrated in two experiments based on the empirical traffic flow data. The results show that the method is able to improve the performances of the short-term traffic forecasting algorithms. It is then suggested to consider the proposed method as a preprocessing procedure in practical forecasting applications.

This study proposes a novel Graph Convolutional Neural Network with Data-driven Graph Filter (GCNN-DDGF) model that can learn hidden heterogeneous pairwise correlations between stations to predict station-level hourly demand in a large-scale bike-sharing network. Two architectures of the GCNN-DDGF model are explored; GCNNreg-DDGF is a regular GCNN-DDGF model which contains the convolution and feedforward blocks, and GCNNrec-DDGF additionally contains a recurrent block from the Long Short-term Memory neural network architecture to capture temporal dependencies in the bike-sharing demand series. Furthermore, four types of GCNN models are proposed whose adjacency matrices are based on various bike-sharing system data, including Spatial Distance matrix (SD), Demand matrix (DE), Average Trip Duration matrix (ATD), and Demand Correlation matrix (DC). These six types of GCNN models and seven other benchmark models are built and compared on a Citi Bike dataset from New York City which includes 272 stations and over 28 million transactions from 2013 to 2016. Results show that the GCNNrec-DDGF performs the best in terms of the Root Mean Square Error, the Mean Absolute Error and the coefficient of determination (R2), followed by the GCNNreg-DDGF. They outperform the other models. Through a more detailed graph network analysis based on the learned DDGF, insights are obtained on the black box of the GCNN-DDGF model. It is found to capture some information similar to details embedded in the SD, DE and DC matrices. More importantly, it also uncovers hidden heterogeneous pairwise correlations between stations that are not revealed by any of those matrices., Comment: 9 figures, 3 tables, accepted by Transportation Research Part C: Emerging Technologies

As a newly-emerging travel mode in the era of mobile internet, ride-hailing that connects passengers with private-car drivers via an online platform has been very popular all over the world. Although it attracts much attention in both practice and theory, the understanding of ride-hailing is still very limited largely because of the lack of related data. For the first time, this paper introduces ride-hailing drivers’ multi-day trip order data and portrays ride-hailing mobility in Beijing, China, from the regional and driver’s perspectives. The analyses from the regional perspective help understand the spatiotemporal flowing of the ride-hailing demand, and those from the driver’s perspective characterize the ride-hailing drivers’ preferences in providing ride-hailing services. A series of findings are obtained, such as the observation of the spatiotemporal rhythm of a city in using ride-hailing services and two categories of ride-hailing drivers in terms of the correlation between the activity space and working time. Those findings contribute to the understanding of ride-hailing activities, the prediction of ride-hailing demand, the modeling of ride-hailing drivers’ preferences, and the management of ride-hailing services.

Taxi is an indispensable mode in the urban public transportation. Although many studies have explored the travel patterns of taxi trips, few have combined taxi and subway to reveal their intermodal relationship. To bridge the gap, this study utilized taxi’s trajectory data to investigate its relationship with subway. Considering the multifaceted relationship between taxi and subway in operation, taxi trips are categorized into three types, namely, subway-competing, subway-extending, and subway-complementing taxi trips. The characteristics of each type of taxi trips reflect the specialties and their interactions with subway. The origin/destination distributions of taxi and subway trips are compared and analyzed. Furthermore, the supply and demand of taxi within the buffer zone of each subway station are analyzed to reflect the difficulty of hailing taxis. The negative binomial regression models are used to explore the relationship between taxi trips and subway ridership. The results show that there is a significantly positive correlation between taxi trips and subway ridership.

This paper proposes a new traffic assignment model, named an N-path logit-based stochastic user equilibrium (NPSUE), which considers not only road users' perceptual errors in path choices, but also users' heterogeneity that allows the numbers of the limited path sets for each user to be different. The property of the N-path user equilibrium (NPUE) proposed in Dung-Ying Lin (2014) is first reanalyzed, and then the optimization program for the NPSUE models as well as the property of the optimal solution are explicitly discussed. For each type of users, the NPSUE assignment results are consistent with the logit-assignment results in their limited path set. Both the NPUE and NPSUE models are tested in three road networks: an simple grid network, Nguyen and Dupuis' network and Sioux Falls network. The numerical results show that (1) the NPUE can be equivalent to the UE under certain conditions, although the users don't have full information on the available paths; (2) the proposed NPSUE may provide an ideal equilibrium state for both system managers and road users; (3) the provision of appropriate path guidance information is beneficial to both system managers and road users; otherwise, the provision of excessive information may decrease the performance of the road network, and (4) under the NPSUE conditions, the provision of excessive information may decrease users' satisfaction.

Loop detectors distributed on freeways are very vulnerable and could be damaged or malfunctioned due to improper sealing, pavement deterioration. This may lead to poor travel time estimation as most of existing methodologies require detailed data collected from numerous detectors along a specified freeway route. To address this problem, this study proposes an effective and reliable methodology for real-time freeway travel time estimation with data from sparse detectors. In contrast to the existing methods, the proposed methodology requires significantly less number of detectors but maintains fairy good performance on travel time estimation. The proposed methodology utilises a self-organised mapping algorithm to cluster the detectors with similar traffic patterns. The data collected from the representative detectors within each cluster is then employed to estimate the travel time based on a support vector regression model. The case studies conducted for three selected freeway routes in Northern California over 3 weeks demonstrate that the proposed methodology accurately captures the fluctuation of travel time induced by the variations of traffic states. The estimated results are exceptionally accurate with smaller mean errors and root-mean-squared errors compared with the benchmark values obtained from the well-known performance measurement system in California.

Automated vehicles (AVs) are widely considered to play a crucial role in future transportation systems because of their speculated capabilities in improving road safety, saving energy consumption, reducing vehicle emissions, increasing road capacity, and stabilizing traffic. To materialize these widely expected potentials of AVs, a sound understanding of AVs’ impacts on traffic flow is essential. Not surprisingly, despite the relatively short history of AVs, there have been numerous studies in the literature focusing on understanding and modeling various aspects of AV-involved traffic flow and significant progresses have already been made. To understand the recent development and ultimately inspire new research ideas on this important topic, this survey systematically and comprehensively reviews the existing AV-involved traffic flow models with different levels of details, and examines the relationship among the design of AV-based driving strategies, the management of transportation systems, and the resulting traffic dynamics. The pros and cons of the existing models and approaches are critically discussed, and future research directions are also provided.

In the era of big data, mining data instead of collecting data are a new challenge for researchers and engineers. In the field of transportation, extracting traffic dynamics from widely existing probe vehicle data is meaningful both in theory and practice. Therefore, this article proposes a simple mapping-to-cells method to construct a spatiotemporal traffic diagram for a freeway network. The method partitions a network region into small square cells and represents a real network inside the region by using the cells. After determining the traffic flow direction pertaining to each cell, the spatiotemporal traffic diagram colored according to traffic speed can be well constructed. By taking the urban freeway in Beijing, China, as a case study, the mapping-to-cells method is validated, and the advantages of the method are demonstrated. The method is simple because it is completely based on the data themselves and without the aid of any additional tool such as Geographic Information System software or a digital map. The method is efficient because it is based on discrete space-space and time-space homogeneous cells that allow us to match the probe data through basic operations of arithmetic. The method helps us understand more about traffic congestion from the probe data, and then aids in carrying out various transportation researches and applications.

An electric vehicle routing problem with charging time and variable travel time is developed to address some operational issues such as range limitation and charging demand. The model is solved by using genetic algorithm to obtain the routes, the vehicle departure time at the depot, and the charging plan. Meanwhile, a dynamic Dijkstra algorithm is applied to find the shortest path between any two adjacent nodes along the routes. To prevent the depletion of all battery power and ensure safe operation in transit, electric vehicles with insufficient battery power can be repeatedly recharged at charging stations. The fluctuations in travel time are implemented to reflect a dynamic traffic environment. In conclusion, a large and realistic case study with a road network in the Beijing urban area is conducted to evaluate the model performance and the solution technology and analyze the results.

Short-term origin-destination (OD) flow prediction in urban rail transit (URT) plays a crucial role in smart and real-time URT operation and management. Different from other short-term traffic forecasting methods, the short-term OD flow prediction possesses three unique characteristics: (1) data availability: real-time OD flow is not available during the prediction; (2) data dimensionality: the dimension of the OD flow is much higher than the cardinality of transportation networks; (3) data sparsity: URT OD flow is spatiotemporally sparse. There is a great need to develop novel OD flow forecasting method that explicitly considers the unique characteristics of the URT system. To this end, a channel-wise attentive split-convolutional neural network (CAS-CNN) is proposed. The proposed model consists of many novel components such as the channel-wise attention mechanism and split CNN. In particular, an inflow/outflow-gated mechanism is innovatively introduced to address the data availability issue. We further originally propose a masked loss function to solve the data dimensionality and data sparsity issues. The model interpretability is also discussed in detail. The CAS-CNN model is tested on two large-scale real-world datasets from Beijing Subway, and it outperforms the rest of benchmarking methods. The proposed model contributes to the development of short-term OD flow prediction, and it also lays the foundations of real-time URT operation and management., Comment: This paper has been accepted by the Transportation Research Part C: Emerging Technologies as a regular paper

This paper investigates the optimal routing design problem of a community shuttle system feeding to metro stations based on demand-responsive service. The solution aims to jointly optimize a set of customized routes and the departure time of each route to provide a flexible shuttle service. Considering a set of on-demand trip requests between bus stops and metro stations, a mixed-integer optimization model is formulated to minimize the total system cost, including the operation cost and passenger&rsquo, s in-vehicle cost, subject to the constraints on the route length, time window, detours, and vehicle capacity. To solve the problem, two metaheuristic algorithms, i.e. a tabu search (TS) and a variable neighborhood search (VNS), with different internal operators are specifically designed. A case study based on a realistic network is conducted to test the model and the solution, and comparisons of the performance of different algorithms are investigated.

Accessibility has drawn extensive attention from city planners and transportation researchers for decades. With the benefits of large-scale and varying time, this study aims to combine the taxi global positioning system (GPS) data with a cumulative opportunity measure to calculate taxi accessibility in Beijing, China. As traffic conditions vary significantly over time and space, we select four typical time periods and introduce a grid-based method to divide the study area into grid cells. Both the GPS signals and opportunities that include the constant points of interest, total drop-offs, and dynamic drop-offs, are aggregated in these grid cells. The cumulative opportunity measure counts all reachable grid cells within the given travel time threshold, along with the corresponding opportunities. The results demonstrate that the accessibility varies in the four time periods, with better performance seen in the late-night hours. Although the spatial distributions of the three kinds of opportunities are different, these accessibilities show great similarity. In addition, the relative accessibilities of different measures are highly correlated. In general, grid cells with higher accessibilities in one time period are likely to also have higher accessibilities in other time periods. Moreover, the results suggest that taxi accessibility can be measured from its trajectory data only.

This paper suggests a probability-dominant user equilibrium (PdUE) model to describe the selfish routing equilibrium in a stochastic traffic network. At PdUE, travel demands are only assigned to the most dominant routes in the same origin-destination pair. A probability-dominant rerouting dynamic model is proposed to explain the behavioral mechanism of PdUE. To facilitate applications, the logit formula of PdUE is developed, of which a well-designed route set is not indispensable and the equivalent varitional inequality formation is simple. Two routing strategies, i.e., the probability-dominant strategy (PDS) and the dominant probability strategy (DPS), are discussed through a hypothetical experiment. It is found that, whether out of insurance or striving for perfection, PDS is a better choice than DPS. For more general cases, the conducted numerical tests lead to the same conclusion. These imply that PdUE (rather than the conventional stochastic user equilibrium) is a desirable selfish routing equilibrium for a stochastic network, given that the probability distributions of travel time are available to travelers.

This paper proposes a convolutional neural network (CNN)-based method that learns traffic as images and predicts large-scale, network-wide traffic speed with a high accuracy. Spatiotemporal traffic dynamics are converted to images describing the time and space relations of traffic flow via a two-dimensional time-space matrix. A CNN is applied to the image following two consecutive steps: abstract traffic feature extraction and network-wide traffic speed prediction. The effectiveness of the proposed method is evaluated by taking two real-world transportation networks, the second ring road and north-east transportation network in Beijing, as examples, and comparing the method with four prevailing algorithms, namely, ordinary least squares, k-nearest neighbors, artificial neural network, and random forest, and three deep learning architectures, namely, stacked autoencoder, recurrent neural network, and long-short-term memory network. The results show that the proposed method outperforms other algorithms by an average accuracy improvement of 42.91% within an acceptable execution time. The CNN can train the model in a reasonable time and, thus, is suitable for large-scale transportation networks.

To alleviate traffic congestion, a variety of route guidance strategies has been proposed for intelligent transportation systems. A number of the strategies are proposed and investigated on a symmetric two-route traffic system over the past decade. To evaluate the strategies in a more general scenario, this paper conducts eight prevalent strategies on a asymmetric two-route traffic network with different slowdown behaviors on alternative routes. The results show that only mean velocity feedback strategy is able to equalize travel time, i.e., approximate user optimality; while the others fail due to incapability of establishing relations between the feedback parameters and travel time. The paper helps better understand these strategies, and suggests mean velocity feedback strategy if the authority intends to achieve user optimality., Comment: 16 pages, 8 figures

This paper proposes a day-to-day route choice model based on travellers’ behavioural characteristics. The model is able to reflect travellers’ route choice in one day and describes the diversity of route choice behaviours. First, characteristics of the day- to-day route choice and travellers’ bounded rationality are analysed. A loop architecture based on the evolution of travellers’ daily choices is presented; deliberate route choice and habitual route choice in the architecture are discussed. A model is then proposed based on the theoretical analysis. In this model, travellers’ behavioural characteristics in the day-to-day route choice are considered; i.e., forgetting previous travel times, risk attitude, habit effect, traveller's amount of attention on the route choice being made, perception of actual travel time, etc. Results of a numerical experiment show that the tendencies of the simulated travellers’ choices are reasonable and similar with that in reality, and that the model is able to reproduce travellers’ route choice with more behavioural characteristics.