輻輳海域シミュレーションシステムによる海域評価

Real time traffic flow data in terms of road-segments traffic densities are relevant information at the basis of several smart services, such as smart routing, smart planning for evacuations, planning of civil works on the city, etc. The traditional methods for traffic flow measured via sensors using data from navigator Apps (e.g., TomTom, Google map, Waze) could be very expensive to be acquired. For this reason, there is the space for low cost and fast solutions for dense traffic flow reconstruction from scattered data coming from a limited number of street sensors spread on the city. The proposed method is based on differential equations and physical constraints applied to a detailed street graph which is enriched of several features. A stochastic learning approach has been adopted to estimate the weights representing in certain sense the road-segments capacity at each time slot of the day. The proposed solution allows computing in real-time the traffic density reconstruction in unmeasured road-segments. The solution has been validated estimating the error in the places where the sensors are positioned, excluding each of them iteratively and reconstructing the flow without it. Then, it has been possible to estimate the error between the reconstructed traffic density and the measured values. This approach allowed setting up a converging algorithm for estimating the traffic density in the whole city graph from detailed parameters. The proposed reconstruction model has been created by exploiting open and real-time data in the context of Sii-Mobility research project by using Km4City infrastructure in the area of Florence, Italy, for its corresponding Smart City solution.

A hybrid short-term traffic flow forecasting method based on spectral analysis and statistical volatility model

This paper presents an adaptive traffic signal controller algorithm that allows real time traffic flow data to control traffic signal timing for an isolated intersection. The proposed design allows greater flexibility and control of signal timings while keeping in view multiple criterions. It is based on simple rules that allow traffic flows on all intersection branches to control the timing of current signal phase without causing unrealistic delays on low volume side streets. The algorithm is also adjusted for capacity of different links. Detector data has been used to assess congestion situation and respond accordingly. A priority scheme for priority vehicles is also proposed. As a proof of concept, the proposed algorithm has been implemented on FPGA as a parameterized VERILOG model allowing adjustment to any intersection size and topology. An evaluator circuit was designed to assess the performance as compared to actuated signal controller. Improvement was observed in total stopped delay, number of stops and queue lengths.

In order to alleviate the current urban traffic congestion pressure and provide accurate and reliable traffic condition information, it is difficult to select the optimal state parameters for the original Hidden Markov model (HMM) and the state number redundancy determined during the training process leads to the model over-provisioning. The problem of weak is integration and generalization. An improved Hidden Markov Model for traffic flow prediction is proposed to more effectively fit the actual urban road intersection traffic flow. In the calculation of the negative log-likelihood function, an Akaike information criterion (AIC) or a Bayesian information criterion (BIC) penalty term is added, and the Baum-Welch algorithm is combined to optimize the optimal state number of the model. Experiments are carried out based on the collected real traffic flow and GPS feature data. The results show that the optimized hidden Markov model is superior to the original model in the accuracy and generalization ability of traffic flow prediction.

In this paper, we propose a Gaussian distribution dynamic freeway segmentation algorithm according to the traffic speed on the freeways. The proposed algorithm adopts a two-stage policy, in the first stage, we divide the freeway into equal length segments via some fixed static standard, e.g. 1000 meters per segment, and we calculate the average speed on each of them. In the second stage, we partition the segment obtained from first stage into two equal parts if the variance of its speeds is larger than predefine max-variance. And then we merge the approximate segments through a variance-based method. Finally, we do some experiments by using the real traffic flow data, the results show our algorithm promise.

Background: In optimal traffic signal timing, some researchers proposed a single objective genetic algorithm to optimize the timing plan at an isolated intersection. However, the genetic algorithm belongs to a natural selection procession. It means that a suggested model might have a local, optimal result instead of global optimization. A few researchers have tried to avoid local optimization values by making many assumptions for the suggested model, these estimations lacked comprehensive theoretical bases in the transportation field. Objective: The objective of this study is to contribute a comprehensive optimization solution, by applying multiple objective genetic algorithms, to minimize the effective green time and cycle length at a complex urban intersection. Methods: First, the fitness function was established by the minimum issues of average control delay and queue length at the complex isolated intersection. Secondly, constraint functions were identified based on a scientific basis to provide a comprehensive hypothetical model. After running the hypothetical model with single and multiple objective genetic algorithms and real traffic flow data, the results were compared between the use of multiple genetic algorithms and the use of a single-objective genetic algorithm, between an existing traffic signal timing plan and a suggested traffic signal timing plan. Then, the traffic simulation model for the complex intersection was generated to validate the effectiveness of the suggested method. Results: After comparison, the suggested model was found to be more efficient than the existing traffic signal timing at the complex intersection. Conclusion: This study demonstrated multiple objective genetic algorithms that overwhelmed the single objective genetic algorithm in optimal traffic signal timing. The multiple objective genetic algorithms could be effectively used to handle traffic optimization at a complex large-scale intersection. Furthermore, a comprehensive solution of applying multiple genetic algorithms to deal with traffic signal optimization has been generated in this research.

<p>In this paper, the characteristics of the potential behavior of ships contained in AIS information are con-sidered, in order to ensure the safety of navigation in merchant ships’ fishing areas and reduce the occur-rence of collisions between commercial fishing boats. Based on the DBSCAN density clustering algorithm, the identification model of busy water in fishing areas is developed and applied to Minnan fishery in the Taiwan Strait. In addition, a real ship simulation is performed based on real traffic flow data. The results demonstrate that the proposed model can accurately identify the spatial distribution and scale of busy fishing area, adjust the algorithm parameters according to the merchant ship scale, and generate the target-ed recommended route decision for the fishing area. Finally, the STELLAR EXPRESS is compared with the original route. The obtained results show that the recommended route provides a high safety in the fishing area, while retaining the ship operating costs. The model results can be used as a reference for maritime security departments in order to divide the fishing area warning water, and the identification method provides a novel approach for the safety supervision of maritime department. </p> <p> </p>

In contemporary traffic flow theory, the distinction between user classes is rarely made. However, we envisage that the accuracy and the descriptive power of the macroscopic traffic flow models can be improved significantly by separating user classes and their specific flow characteristics. As a consequence, the possibility of improved estimation and prediction of traffic flow conditions becomes available. Additionally, the availability of a realistic multiple user-class traffic flow model enables the automated generation of user-dedicated traffic control policies by means of mathematical optimal control theory. A macroscopic multiple user-class model is derived from mesoscopic principles. In opposition to earlier multiple user-class models, the model presented here implicitly defines equilibrium relationships between traffic density and equilibrium velocities as a function of current traffic conditions, traffic composition, and distribution of user-class-dependent desired velocities. Additionally, the velocity variance variable is introduced describing deviations from the average speed within the user classes. The multiple user-class model identifies competing processes. On the one hand, drivers attempt to traverse the freeway at their desired velocity; on the other hand, they adjust their velocity because of interaction with slower vehicles. These processes can result in self-formation of apparently random local structures. Finally, the proposed model satisfies the anisotropy condition and the invariant personality condition.

In satellite video vehicle tracking, due to the tracking failure and tracking loss caused by similar characteristics of the target and obstacle occlusion, respectively, the traffic flow extraction accuracy is reduced. To address these issues, an improved traffic flow extraction method for satellite video based on kernelized correlation filter (KCF) was proposed. First, we introduced a multifeature fusion strategy into the KCF based on the discrete Fourier transform (DFT) framework to enhance vehicle tracking accuracy and reduce tracking drift and jumps. Second, we utilized the Kalman filter for trajectory prediction to reduce the loss of target during vehicle tracking. Compared with other mainstream algorithms on the satellite video dataset, the results showed that the tracking accuracy and success rate of the proposed method reached 86.74% and 79.96%, respectively. Finally, the virtual detection line method was used to extract the traffic flow. The experimental results showed that compared with the real traffic flow data obtained by visual method, the accuracy of satellite video traffic flow extraction by virtual detection line was 98.48% under noncongestion condition and 90.18% under congestion condition.

In large cities, a situation often arises when roundabout intersections of highways no longer cope with the passage of traffic flows. The main reasons for this problem are the inconsistency of the potential traffic capacity of the roundabout with real traffic flows, as well as the lack of knowledge and skills of drivers on the roundabout carriageway. To reduce the influence of the human factor, various solutions are proposed, but, as practice has shown, ducted roundabouts or «turbo rings» are the most effective. A feature of this configuration of the carriageway is the predictability of driver behavior due to the physical separation of lanes for traffic. In a number of countries, turbo ring intersections have already been put into operation, which made it possible to confirm the alleged advantages of this configuration: the absence of changeover maneuvers on the ring carriageway, the need to give way to a maximum of two traffic flows, and a decrease in the average travel speed along the ring intersection. The aim of this study was to compare the performance of the existing threelane intersection, characterized by a predominance of lefthand turn flow during rush hours, and the developed design of the turbo ring interchange. Empirical data on traffic flows and length of queues, as well as a matrix of correspondences, were used to calibrate the computer micromodel. Micromodeling was carried out by means of: specifying the matrix of correspondences, taking into account the distribution of the traffic flow in time; setting the rules of priority, speed of approach to the intersection and speed reduction zones; development of scenarios for the analysis of morning and evening rush hours, fluctuations in traffic flows in time. Performance comparisons were made for the node as a whole, as well as for the selected busiest routes in terms of average and maximum congestion lengths, average latency and downtime. To assess the performance of the intersections, it was proposed to use foreign methods for assessing the capacity based on the mathematical models of Bowie, Hagring and Fortine. It was revealed that the transformation of the existing circular intersection into a turbo ring made it possible not only to increase the throughput of the node, but also to reduce congestion at the entrances and exits, which determines the practical significance of the study. To study the possibility of integrating the methodology for assessing the throughput, considered in the article, into the domestic method for choosing the type of intersections, which is the basis of regulatory and technical documents. The ultimate goal of the study is to create a new methodology for assessing the throughput of circular intersections of various types, including taking into account the peculiarities of the functioning of circular intersections on the territory of the Russian Federation.

Phase diagram in multi-phase heterogeneous traffic flow model integrating the perceptual range difference under human-driven and connected vehicles environment

Real-virtual consistent traffic flow interaction

The study of traffic flow characteristics is essential for designing highway systems for which different traffic influencing variables are of primary importance. The study aims to assess the impact of a range of traffic flow and density levels on the spatio-temporal (lateral placement and time headway) traffic characteristics of highways. A substantial amount of traffic data like speed, time headway (TH), and lateral placement (LP) of vehicles was collected continuously for 12 h on six Indian highway sections. The DPCU for different vehicles with different flow and density rates were calculated. The speed and TH descriptive statistics and probability distribution functions at varying flow and density levels were also analyzed. To determine the most appropriate probability distribution function, the goodness of fit test was used. Furthermore, the impact of traffic speed, flow, and density on the LP of vehicles under different traffic conditions was assessed using descriptive statistics. It was observed that the traffic flow and density levels had an unusual effect on the DPCUs of vehicles. The mean traffic speed under high flow and density decreased by 33.3 and 18.3%, while the mean TH under high flow and density decreased by 28.4 and 29.9%. Also, the results indicated that traffic speed and TH data exhibit a different distribution of probability functions depending on the traffic flow and density. The descriptive statistics on the LP of vehicles show the existence of a significant difference of the same concerning the different ranges of traffic speed, flow, and density.KeywordsSpeedFlowDensityTime headwayLateral placementMixed traffic

In recent studies on traffic flow, cellular automata (CA) have been efficiently applied for simulating the motion of vehicles Since each vehicle has an exclusion volume and moves by itself not being ruled by the Newton's laws of motion, CA is quite suitable for modelling traffic flow In the present paper, we propose a stochastic CA model for traffic flow and show the availability of CA modelling for the complex phenomena that occur in real traffic flow.

This paper proposes a methodology for collision prevention in car following scenarios. For this purpose, Emotional Learning Fuzzy Inference System (ELFIS) approach is used to simulate and predict the behavior of a driver-vehicle-unit in a short time horizon ahead in the future. Velocity of the follower vehicle and relative distance between the follower and the lead vehicles are predicted in a parallel structure. Performance of the proposed algorithm is assessed using real traffic data and superior accuracy of this method is demonstrated through comparisons with another available technique (ANFIS). The predicted future driving states are then used to judge about safety of the current driving pattern. The algorithm is used to generate a warning message while a safe-distance keeping measure is violated in order to prevent a collision. Satisfactory performance of the proposed method is demonstrated through simulations using real traffic data. The proposed method can be applied, in real time, for a variety of applications including driver assistant and collision prevention systems as well as other intelligent transportation applications. Language: en