Search

Traffic arrival pattern is an indispensable part of the traffic state analysis and optimization of signal control schemes at the urban intersection. In some previous research relevant to the traffic parameters at the intersection, such as the queue length estimation, delay estimation, and vehicle trajectory reconstruction, the vehicle arrival patterns are assumed to obey a Normal or Poisson distribution. However, the actual traffic arrival patterns at the intersection are dynamic and more complex. License plate recognition (LPR) data containing rich vehicle trajectory information are promising in data-driven applications of traffic control and management. We propose a framework based on the LPR data, to infer the dynamic traffic arrival pattern at the intersection. The proposed integrative data-driven framework consists of three main parts, a method for inferring traffic signal cycle parameters, deriving the cycle start time and cycle length from time headway sequences; a link travel time estimation method, extracting the valid travel time at the selected time interval, and an arrival pattern estimation model based on the above modules to perform the second-level estimation at the downstream intersection. The unique characteristics of LPR data are exploited and utilized in this study. To validate the proposed method, the LPR data from two adjacent intersections in the city of Changsha in China are used. Numerical results reveal that the proposed framework can achieve satisfactory estimation under different traffic scenarios. The findings in this study can be extended for supporting efficient traffic control applications. • A statistical framework to infer the fine-grained traffic arrival pattern is proposed. • The signal cycle parameters are inferred by exploiting the features of the time headway sequences. • The valid travel time is extracted at the selected time interval based on the Gaussian mixed model. • The proposed framework could achieve satisfactory performance in different traffic scenarios. [ABSTRACT FROM AUTHOR]

The explicit measurement of the evolution process of queues during a cycle has a special significance for evaluating the control scheme at signalized intersections, especially when a certain unconventional operation or intervention is conducted. In order to measure the dynamic behavior of queues at signalized intersections with waiting areas, this paper develops a stochastic queueing model to analytically simulate the evolution process of queues. The key queueing performance indices such as the statistical distributions of queue length and mean values of queue length along the time are derived. Using these indices, the queueing process of vehicles at the intersection with waiting areas can be explicitly formulated, in particular, the dynamic behavior of queues along the time during a cycle can be captured, and the formation of maximum queue length during a cycle can also be measured. Model validation shows that the proposed queueing model can provide an accurate estimation of queueing performance of signalized intersections with waiting areas. In addition, the effects of the use of waiting areas on the evolution process of queue length during a cycle are also tested by numerical simulations under different conditions. [ABSTRACT FROM AUTHOR]

Industrial loop detectors (ILDs) are the most common traffic detectors. In Shanghai, most of the ILDs are installed in a single loop way, which can detect various parameters, such as flow, saturation, and so on. However, they cannot detect the speed directly, which is one of the key inputs of intelligent transportation systems (ITS) for identifying the traffic state. Thus, this paper is dedicated to estimate speed accurately. It proposes a new algorithm that multiple kernel support vector regression (MKL-SVR) to complete this goal, which improves the accuracy and robustness of the speed estimation. Extensive experiments have been performed to evaluate the performances of MKL-SVR, compared with polynomial fitting, BP neural networks and SVR. All results indicate that the performances of MKL-SVR are the best and most robust. [ABSTRACT FROM AUTHOR]

Realistic modeling traffic flow dynamics in urban traffic network remains a big challenge at present due to the complex nonlinear characteristics of traffic flow. In this paper, a novel and model-free iterative learning control (ILC) strategy for balancing traffic flow in urban traffic network is proposed. To tackle the randomly varying trial lengths in the iteration domain of traffic system, an iterative-average operator is introduced in the proposed ILC law for tracking tasks with non-uniform trial lengths, which thus mitigates the requirement on classic ILC that all trial lengths must be identical. The learning convergence condition of the ILC strategy in iteration-average and expectation is derived through rigorous analysis. The performance and the effectiveness of the ILC strategy are analyzed by simulations on a test road network. The results show that the proposed ILC strategy can homogeneously balance the accumulation in the network and improve the network mobility. [ABSTRACT FROM AUTHOR]

Urban signal intersection as the key node in the urban road network, its operational efficiency directly affects the overall operational efficiency of the road network traffic. The existing urban road signal intersections cannot be broadened due to the limitation of road size and boundary lines of road. Some signalized intersections adopt straight-right lane design to improve the traffic efficiency of vehicles at intersections. This design can improve the traffic efficiency of straight-through vehicles during the green light phase. However, straight-through vehicles stopping in the straight-right lane during the red light will prevent right-turning vehicles from passing the intersection. This phenomenon causes delays for right-turning vehicles. In order to deal with the above issues, this paper takes the concept of separating straight and right-turning vehicles in spatial and temporal on the dynamic straight-right lane (DSRL). The organization form of vehicles entering DSRL is designed based on signal phase time. The control strategy for DSRL with the interaction of traffic lights and pre-signal is formulated integrated with the intersection design. The delay indexes of straight and right-turning vehicles under DSRL design are quantified using cellular automata model to calculate vehicle delays under the design form of DSRL. Finally, the test platform based on the actual intersection is established with MATLAB to simulate the operation of the vehicles under the DSRL design. The experiment shows that the delay reduction degree of straight-through vehicles is between 1.2% and 4.5%, and the reduction degree of right-turning vehicles delay is more than 66%, that the operation efficiency of intersection is improved. • In terms of design concept, this paper proposes DSRL design scheme, which separates straight-through vehicles and right-turning vehicles from temporal and spatial. Right-turning vehicles are not delayed during the red phase while both straight and right-turning vehicles can successfully pass the intersection during the green phase under DSRL. The traffic efficiency of vehicles at the intersection can be greatly improved. • In terms of control method, the control strategy proposed in this paper is more flexible than the original variable lanes. This paper proposes pre-signal and traffic signal control scheme under DSRL. The calculation method of signal timing parameters and suggesting parameters of signal timing are given according to the ideal operating conditions of the traffic flow. The signal control strategy and DSRL design cooperating together can make traffic flow operate better and fully exploit the lane traffic capacity. • In terms of test results, this paper builds a microscopic model to evaluate the performance of the DSRL. By systematically analyzing key parameters, DSRL design can effectively reduce vehicle delays in different traffic environments. This paper also lays the foundation for optimizing the indicators under DSRL design and provides guidance for the implementation of DSRL design. [ABSTRACT FROM AUTHOR]

Abstract In this paper, we study the application of perimeter flow control by simulation of cellular automaton model on urban traffic system. We find that the relation of traffic flow and vehicle density (Macroscopic Fundamental Diagram, MFD) will have different shapes for the core area and the peripheral area. The MFD shows free-flow state, saturate flow state and congestion state. But the magnitude of maximal flow and the position for congestion transition are different for the whole system and the core area. We suggest to realize the perimeter control strategy by assigning a special prohibiting phase to the perimeter traffic lights for the roads entering the core area. The strategy is controlled by two critical densities ρ 1 and ρ 2 , which are determined by the MFD of the core area. Simulations show that both the average arrival rate and the average flow will be greatly improved with the perimeter flow control strategy. In addition, the perimeter flow control strategy can increase the critical density of traffic congestion. The probability of system-wide gridlock will decrease, and the system can perform well under both close and open boundary conditions. All the results indicate that the perimeter flow control strategy can effectively improve the performance of the traffic system. Highlights • MFD has different shape for the system and for the core area. • Parameters of perimeter control should be determined by the MFD of core area. • Perimeter control can reduce gridlock and increase the critical density. [ABSTRACT FROM AUTHOR]

Abstract Based on the cellular automaton model and the concept of ant pheromone, this paper proposes a new route guidance strategy, called the ant pheromone route guidance strategy, in which the vehicles are regarded as special types of ants and their traffic information is regarded as the ant pheromone. To evaluate its performance, the new route guidance strategy was applied and compared with other three typical route guidance strategies under three different route scenarios, respectively, with open boundary conditions based on the Nagel–Schreckenberg cellular automaton model. First, in a symmetrical two-route scenario with two exits, results showed that the new route guidance strategy and vacancy length route guidance strategy were optimal. They outperformed the other strategies in terms of the value, stability, and balance of vehicle number, and the average speed and average flux on each route. To understand the impact of the strategy on traffic states, flux–density evolution process was studied. In addition, the influence of evaporation rate was also investigated. Second, in an asymmetrical two-route scenario with one exit, the simulation results also proved that ant pheromone route guidance strategy and vacancy length route guidance strategy were the best. Finally, in a symmetrical two-route scenario with a traffic accident, we compared the new route guidance strategy with the vacancy length route guidance strategy. Results indicated that the new route guidance strategy has distinct advantages. In short, in all three route scenarios we investigated, the ant pheromone route guidance strategy performed best. Highlights • This paper proposes the ant pheromone route guidance strategy based on CA model. • The new strategy takes vehicles as ants and vehicles' data as ant pheromones. • Adopting the NS model, this strategy is applied to three two-route scenarios. • Results indicate that the new strategy outperforms other route guidance strategies. • In traffic accident, the new strategy can reduce congestion and save recovery time. [ABSTRACT FROM AUTHOR]

Abstract By means of the Nagel–Schreckenberg model, we have investigated into the maximum vehicular flow rate of traffic processing bottlenecks. The evaluated analytical form of this flow rate is found to give quantitative insights into the underlying physics of collective vehicular motions constrained by these bottlenecks. Our analysis shows that for large-scale expansion, a new class of processing bottleneck known as the serial bottleneck is more efficient than the conventional parallel bottleneck in the absence of human driving behavior. When characteristics such as slow-to-start is considered in the model, the consequential delay due to human reaction time not only degrade the overall efficiency, it also diminishes the efficacy of serial processing such that a serial bottleneck is no longer tenable for traffic processing. These results point to the fundamental importance of optimizing traffic efficiency, which we illustrate by elucidating the detailed mechanisms with which vehicles interact collectively in the bottlenecks. In particular, we demonstrate that by constructing combinations of serial and parallel bottlenecks, optimal efficiencies are achieved via configurations with few (many) lanes of a large (small) number of serial units when the processing time is short (long). A direct implication of these results is that autonomous self-driving vehicles could serve to improve the transportation capacity for the densely populated urban cities of the future, due to the intrinsically more efficient collective vehicular motions through these bottlenecks. Highlights • Capacity of serial and parallel bottlenecks have been compared using full analytical solutions. • Analysis shows that the difference in the constrained motion results in the difference in capacity. • Novel combination of serial and parallel expansions was proposed. • The efficacy of these bottlenecks in relevance of real-world traffic have been discussed. [ABSTRACT FROM AUTHOR]

Transitions between two lanes often have a significant impact on various forms of road traffic. To address this problem, we have developed a two-lane asymmetric simple exclusion process model and two hypothetical traffic control strategies, to simulate a futuristic scenario where the timing and location of transitions between two lanes are highly controlled. Various scenarios were proposed to study the effectiveness of these control strategies. An optimized control strategy, whose parameters were determined through an optimization algorithm, is confirmed to effectively maximize the average traffic flow. Consequently, we may identify suitable road sections and the corresponding timings of transitions to resolve congestion in this model. • Regulated transitions between two lanes can improve traffic flow. • Regulations are simulated using the ASEP model and a hypothetical control strategy. • Parameters of the control strategy are determined with an optimization algorithm. • The control strategy can determine positions and timings suitable for transitions. [ABSTRACT FROM AUTHOR]

Traffic flow prediction is an important part of the intelligent transportation system, and accurate traffic flow prediction can better perform the data management and control of the urban road network. Traditional methods often ignore the interaction between traffic flow factors and the spatiotemporal dependence of traffic networks. In this paper, a spatiotemporal multi-head graph attention network (ST-MGAT) is proposed for the traffic flow prediction task. In the input layer, multiple traffic flow variables are used as input to learn the nonlinearity and complexity existing in it. In terms of modeling, the structure of linear gating units is transformed using the full volume and applied to correlation learning in the temporal dimension. The spatial dimension features are then captured using a multi-head attention map network. Ablation experiments are designed to verify the use effect of each module. At the same time, several sets of parameter experiments are designed to obtain the best model parameters. The experimental results show that compared with the baselines, the RMSE of the ST-MGAT model is decreased by almost 2.340% ∼ 22.471%, and the MAE is decreased by almost 7.417% ∼ 26.142%. It is proved that the ST-MGAT model has high performance on the traffic flow prediction task. • A spatio-temporal multi-head graph attention network (ST-MGAT) is proposed. • Transportation network is modeled as a directed graph, and the multi-head graph attention network is used to capture the characteristics of the spatial dimension. • Full convolution structure and gated linear units are used to replace the recurrent neural network, in order to capture the characteristics of the time dimension. • Traffic flow, traffic speed and time occupancy rate are simultaneously used as the input of the model. [ABSTRACT FROM AUTHOR]

Missing and suspicious traffic data is a major problem for intelligent transportation system, which adversely affects a diverse variety of transportation applications. Several missing traffic data imputation methods had been proposed in the last decade. It is still an open problem of how to make full use of spatial information from upstream/downstream detectors to improve imputing performance. In this paper, a tensor based method considering the full spatial–temporal information of traffic flow, is proposed to fuse the traffic flow data from multiple detecting locations. The traffic flow data is reconstructed in a 4-way tensor pattern, and the low-n-rank tensor completion algorithm is applied to impute missing data. This novel approach not only fully utilizes the spatial information from neighboring locations, but also can impute missing data in different locations under a unified framework. Experiments demonstrate that the proposed method achieves a better imputation performance than the method without spatial information. The experimental results show that the proposed method can address the extreme case where the data of a long period of one or several weeks are completely missing. [ABSTRACT FROM AUTHOR]

Accurate traffic state (i.e., flow, speed, density, etc.) on an urban road network is important information for urban traffic control and management strategies. However, due to the limitation of detector installation cost, it is difficult to obtain accurate traffic states through detectors in the whole urban road network with limited detector equipment. In this paper, we review the studies that focus on the missing traffic state estimation problem, especially for the traffic state estimation on the segments without detectors. We provide a way to summarize for readers who have an interest in the different modelling and application of missing traffic state estimation. We first divide the existing studies into three categories: estimation under different missing scenarios, estimation with multi-source data, estimation by fusing different detector types. Then, we summary some existing challenges by the different missing scenarios, data applications, and methodologies. Finally, this work also discusses some future research directions. [ABSTRACT FROM AUTHOR]

Abstract: The analysis of dynamics in traffic flow is an important step to achieve advanced traffic management and control in Intelligent Transportation System (ITS). Complexity and periodicity are definitely two fundamental properties in traffic dynamics. In this study, we first measure the complexity of traffic flow data by Lempel–Ziv algorithm at different temporal scales, and the data are collected from loop detectors on freeway. Second, to obtain more insight into the complexity and periodicity in traffic time series, we then construct complex networks from traffic time series by considering each day as a cycle and each cycle as a single node. The optimal threshold value of complex networks is estimated by the distribution of density and its derivative. In addition, the complex networks are subsequently analyzed in terms of some statistical properties, such as average path length, clustering coefficient, density, average degree and betweenness. Finally, take 2 min aggregation data as example, we use the correlation coefficient matrix, adjacent matrix and closeness to exploit the periodicity of weekdays and weekends in traffic flow data. The findings in this paper indicate that complex network is a practical tool for exploring dynamics in traffic time series. [Copyright &y& Elsevier]

Abstract: In the intelligent transportation system, an effective feedback strategy is of crucial importance to the improvement of traffic condition and transport capability. Based on the seven previously introduced feedback strategies, a new one is introduced, called vacancy length feedback strategy (VLFS). The simulation results in the symmetrical two-route scenario with two exits suggest that VLFS is the optimal one among all the feedback strategies. It outperforms others in terms of the value, stability, average flux and balance of the vehicle number, and also exceeds others for the convenience of its application in the real traffic condition. The later simulation results in the asymmetrical two-route scenario with one exit also prove that VLFS is the best. [Copyright &y& Elsevier]