Applied Mechanics and Materials Vols. 97-98

Abstract: Short-term traffic flow forecasting has a rich and substantive research history. For layered hybrid forecasting approaches, traffic flow series are classified into distinct regimes based on characteristics of different traffic conditions so that regime-specific forecasting models can be trained to adequately and consistently approximate dynamic traffic dynamic behaviors. However, traffic regimes are inherently changing and transient, making it a challenging task to perform regime-specific forecasting. In this paper, an ensemble learning approach is proposed and regime-based forecasting models are developed to address the nonlinearity and non-stationarity of traffic flow data. Case study results demonstrate that the proposed approach can effectively account for transient regimes and produce accurate online short-term traffic forecasts.

Abstract: Aiming at the fact that target ships of ship handling simulator have not the ability of avoiding other ships according to the COLREG 1972, the paper puts forward to add Personifying Intelligent Decision-making for Vessel Collision Avoidance, for short PIDVCA, and vessel automatic control algorithm into solver procedure of target ship, designed and realized target ships to sail according to the COLREG 1972, which make ship handling simulator more realistic simulate the actual navigation environment at sea. Finally, the paper gives the simulation pictures of 2 and 4 target ships intelligent collision.

Abstract: The study on driver’s intention recognition is useful to improve the conclusion accuracy of the risk assessment in current vehicle driving safety precaution system. DS evidence theory is used to study a recognition method for driver’s intention. In order to improve the accuracy of driver’s intention recognition, the driving sequential actions and the driver’s characteristics are also taken into account. The detailed steps of the recognition method for driver’s intention are presented. And the experimental result shows that the recognition method is feasible and effective.

Abstract: With the rapid development of China’s urbanization, the traffic become much more heavy in the city, transportation management problem becomes increasingly serious and city intelligent transportation management command system (CITMCS) has become one of the urgent research and solving subjects of ITMS (intelligent transportation management system). This paper presents a city intelligent transportation management command system. It is an integrated intelligence system of the transportation detection, control, management, decision making, command and communication. It can control the transportation facilities conveniently, provides useful information for transportation user, and ensures travelers to go through the road conveniently and safely. The system will have an extensive prospect of application and development. The paper firstly introduces the concept of the CITMCS, then presents the basic composition and function structure diagram of CITMCS based on information processing and information share, and finally gives the analysis of the relative key technique in system construction process.

Abstract: Simplified Monte Carlo collision model of real-time traffic flow prediction is proposed. In this model, two different road cells, roads and crosses, are configured. Vehicle distribution is generated by real-time traffic flow randomly. Based on two-dimensional topology, Monte Carlo collision between road and vehicles promote time evolution of the system. Monte Carlo collision is the core of the model and traffic flow is study target. The solution of relationship equations of road and vehicles is very simple in this model to speed up the computing. In addition, parameters can be corrected and configured at any time in the process of time evolution. Experimental results show that the model has the advantages of real-time, visual interface, easy configuration, and can be corrected by real-time feedback. The model can not only simulate and predict macroscopic data, such as flow, velocity distribution, but also follow the track of each vehicle in detail. So, the model can be used in researching both macroscopic and microscopic characteristics of vehicle movement.

Abstract: Traffic problem is one of serious problem in our society. Put forward the concept of intelligent transportation systems, is seeking to address traffic problems, and gradually become a research hotspot in traffic areas. As two subsystems, traffic control and traffic guidance systems tend to work independently, this paper aims to establish a coordination with the integration of the two systems, while use the agent technologies which emergy in artificial intelligence to establish an integrated model.

Abstract: In this paper we investigated the stability of the traffic flow on a single lane gradient (uphill/downhill) highway. The linear stability theory was used to analyze the model and get the stability conditions. The theoretical result shows that the slope of the gradient has an influence on the stability of traffic flow. Simulations are carried out to check the slope effect of the traffic flow model. Numerical result is in good agreement with the real traffic situations.

Abstract: Efficient dynamic shortest path algorithm in static networks plays an important role in ITS. To solve this problem, this paper brings forward the dynamic form of Consistency Assumption and Dynamic A* algorithm (A* algorithm based on dynamic lower bound, DA* algorithm) based on dynamic lower bound. DA* algorithm and the dynamic form of Consistency Assumption are described in detail. It is proved that DA* algorithm can solve one origin node to one destination node shortest paths problem in dynamic networks, if DA* algorithm’s dynamic lower bound satisfies the dynamic form of Consistency Assumption.

Abstract: In this work, we apply fuzzy logic method to address the traffic density control problem in a macroscopic level freeway environment with ramp metering. The second-order traffic flow model is firstly formulated. Then traffic density is selected as the control variable in place of traffic occupancy. Based on the traffic flow model and in conjunction with nonlinear feedback theory, a fuzzy logic based traffic density controller is designed. The ramp metering rate is determined by the fuzzy control according to density tracking error and error variation. Triangle curves are used for the membership functions of the fuzzy variables. The rule base including 56 fuzzy rules is also established. Finally, the fuzzy logic based feedback controller is simulated in Matlab software. Simulation results show that this method has good dynamic and steady-state performance, and can achieve an almost perfect tracking performance.

Abstract: According to the traffic conditions in the typical freeway tunnel group in China, an artificial neural network model is constructed for the purpose of predicting the operating speed in freeway tunnel group in this paper. In this model, some input variables are selected from four aspects, including time factors, traffic dynamic factors, road conditions and tunnel environment, and the output variable is the operating speed. Then the sensitivity analysis method is selected to study the effects of input variables on output variable. The results show that this algorithm can avoid the difficulty of constructing traffic flow model comparing to the traditional algorithm, and it is suitable to realize online modeling for speed limit of freeway tunnel group. Results of this research are practical and effective, and it may provide a theoretical foundation for speed limit of freeway tunnel group.