Archives Search

Abstract

Trip demands and patterns on weekends have been changed significantly due to increase in income levels and spread of the 5-day workweek. The increased weekend trips for shopping, leisure, entertainment and social activities have exacerbated traffic congestion in main high...Lead moreTrip demands and patterns on weekends have been changed significantly due to increase in income levels and spread of the 5-day workweek. The increased weekend trips for shopping, leisure, entertainment and social activities have exacerbated traffic congestion in main highways or national roads from Friday afternoons through Sundays. Therefore, it is necessary to activate studies on demand estimate and traffic policy for weekend trips by investigating the specific characteristics of the trips. Former researches have focused on simple analysis of traits of weekend trips and comparison of travel characteristics between weekdays and weekends. This study analyzes the differences between weekday and weekend trips via a statistical technique to derive multiple types of characteristics of weekend trips, and constructs a Tobit model to come up with key factors that may affect the number of trips, using Seoul city’s weekend trip survey data. As results, different estimates of Tobit model for respective Saturdays and Sundays and for weekdays are comparatively shown.

Abstract

This paper proposes an algorithm for labeling and segmenting blue traffic signs indicating directions. In order to discriminate blue signs from blue sky background, opponent color filters, which are found in our visual system, and seeded region growing algorithm are emplo...Lead moreThis paper proposes an algorithm for labeling and segmenting blue traffic signs indicating directions. In order to discriminate blue signs from blue sky background, opponent color filters, which are found in our visual system, and seeded region growing algorithm are employed. Since blue signs are occasionally segmented as several disconnected parts, a connecting algorithm for fragmented signs is devised. After segmentation, a verification step takes place to test if the segmented blue object is a sign candidate or not. After that white objects within the blue object are segmented. Finally the white objects are classified by matching them to templates of arrow patterns according to geometrical features. If no arrow pattern is matched, they are rejected. Classification is done using a decision tree. Processing time was about 100-160 ms/frame with 2.20GHz notebook PC. The system could segment all blue signs from blue sky background. Precision rate of recognition was 71% and recall rate was 94%.

Abstract

This paper describes the city planning support system with using the information generated in the operation log data of On-demand Bus service. In general, it is difficult to get citizens’ mobility data. By analyzing the log data of On-demand Bus, the mobility of persons i...Lead moreThis paper describes the city planning support system with using the information generated in the operation log data of On-demand Bus service. In general, it is difficult to get citizens’ mobility data. By analyzing the log data of On-demand Bus, the mobility of persons in the city will become apparent. Such information is very useful when considering locations of facilities. Developed system can calculate the most appropriate facility location with the On-demand Bus log data and the system is evaluated by the field test in Kashiwa City, Chiba Prefecture. If the log data have a bias, the results of facility locations will become not fair. For example, the age distribution of the On-demand Bus enrollees is different from that of whole people in Kashiwa City. To correct the bias of age distribution, the ratio of each age group was adjusted by weighting the demands. The calculated optimal sites can become more fairly, by adjusting the population rate of the On-demand Bus users to that of the whole field test area.

Abstract

As Green House Gas (GHG) emission has continuously increased over the recent decades, global warming has become one of the critical environmental and societal issues. GHG tax policy is considered as a countermeasure to reduce GHG emission in many countries including Korea...Lead moreAs Green House Gas (GHG) emission has continuously increased over the recent decades, global warming has become one of the critical environmental and societal issues. GHG tax policy is considered as a countermeasure to reduce GHG emission in many countries including Korea. Thus, quantitative understandings of sequential effects of GHG tax on GHG emission are in need for successful implementation. The paper evaluates the impacts of GHG tax on reducing the amount of GHG emissions by shifting motor vehicle users to public transportation modes. Mixed multinomial logit model was used to estimate travelers’ mode choice behavior based on stated preference survey data collected in the Seoul metropolitan areas, Korea. The estimated model was then used to emulate travelers’ mode shift due to imposing GHG tax for a coupled of selected corridors based on revealed preference survey data. This procedure quantitatively forecasts the outcomes of GHG tax in reducing GHG emission and, by doing so, reaches to the useful implications for GHG policy and transportation planning.

Abstract

Traffic count data is the most fundamental information required for rural road planning and evaluation; rural road network evaluation, road traffic management, tolling, and pavement maintenance management are impossible without traffic count data. Because of widespread ru...Lead moreTraffic count data is the most fundamental information required for rural road planning and evaluation; rural road network evaluation, road traffic management, tolling, and pavement maintenance management are impossible without traffic count data. Because of widespread rural road network in both developed and developing countries, gathering permanent traffic data for all roads is not practically possible or even necessary. Thus, a traffic counting program in which both permanent and short-term traffic counts are used to collect traffic volume data is needed. The main aim of this comprehensive program is to obtain the most coverage of the traffic count data with optimum application of sources, including experts and equipment. In this paper, a comprehensive traffic counting program (CTCP) will be developed based on an integer linear programming (ILP) model. The ILP method uses annual average daily traffic (AADT) and monthly traffic standard deviation (MTSD) as the inputs and determines which roads should be surveyed permanently or over the short term. The CTCP specifies the frequencies of short-term traffic counts on selected road. It also applies a clustering method to compute monthly road traffic variation factors in order to infer roads’ permanent traffic data from short term counting stations. The CTCP is executed in Qazvin, one of cities in Iran, to demonstrate the model implementation process. The results show appropriate distributions of permanent and short-term traffic counting stations all over Qazvin’s roads.

Abstract

Accessibility constraints due to traffic congestion resulting from vehicular growth is a major transportation related problem in urban areas. Increasing road infrastructure to reduce traffic congestion is a highly expensive proposition. Controlling traffic signal timing i...Lead moreAccessibility constraints due to traffic congestion resulting from vehicular growth is a major transportation related problem in urban areas. Increasing road infrastructure to reduce traffic congestion is a highly expensive proposition. Controlling traffic signal timing in an optimal condition is undoubtedly one of the most cost-effective means of improving mobility in urban traffic systems. But the high heterogeneity and low level of traffic discipline on urban roads in developing countries like India poses a great challenge in achieving this goal. The solution becomes more complex when the heterogeneity is augmented with poor lane discipline. Coordinated Vehicle Actuated (VA) traffic signal controllers offer better signal time management in a traffic signal network, especially when the traffic is not saturated. Classical method of computing phase-time and cycle-length is an offline process. But the vehicle actuated traffic control demands for real-time computation of signal timings. Deriving the parameters required to compute signal timing in real-time is a complex task in heterogeneous traffic, due to various reasons. Other prominent issues in implementation of ATCS in developing countries are (1) availability of funds and (2) communication overheads. This paper describes an appropriate solution for the effective control of road traffic network having high heterogeneity and poor lane discipline; and also addresses the issues of funds availability by introducing high scalability to the system and optimized communication strategy. This is currently implemented in Indian cities Pune and Jaipur, and being implemented in Kolkata and Ahmedabad. The solution resulted in minimum network delay and improved mobility.

Abstract

We proposed a new method of compressing/de-compressing a moving picture using an I-frame for the streaming video service. When an I-frame, in which a scene is changed, is similar to a previously used frame, the I-frame tag is compressed, instead of the I-frame, to increas...Lead moreWe proposed a new method of compressing/de-compressing a moving picture using an I-frame for the streaming video service. When an I-frame, in which a scene is changed, is similar to a previously used frame, the I-frame tag is compressed, instead of the I-frame, to increase the compression rate. The experiment results are compared with conventional GOP structure and the performance is good.

Abstract

With the deployment of Intelligent Transportation Systems (ITS) infrastructure, the need for long-range traffic flow forecasting is demanding. This paper investigated the performance of a combination method for long-range freeway traffic forecasting, and it was compared t...Lead moreWith the deployment of Intelligent Transportation Systems (ITS) infrastructure, the need for long-range traffic flow forecasting is demanding. This paper investigated the performance of a combination method for long-range freeway traffic forecasting, and it was compared to other statistical models using two statistical error estimates. Results showed that the performance of the models was different with respect to variability of traffic flow. In general, the exponential smoothing and the autoregressive model gave good results for short-term forecasting, but the combination method produced the good results for long-range forecasting as to different traffic conditions.

Abstract

Generally traffic engineering involves traffic data collection for statistics, traffic monitoring, traffic enforcement such as speed and weight limit enforcement and traffic management. Currently many kinds of systems or devices are available which measure or monitor or e...Lead moreGenerally traffic engineering involves traffic data collection for statistics, traffic monitoring, traffic enforcement such as speed and weight limit enforcement and traffic management. Currently many kinds of systems or devices are available which measure or monitor or enforce traffic dedicatedly. The use of these dedicated or nonintegrated systems or devices in traffic data-collection or monitoring and traffic enforcement create a lot of problems in terms of data consistency, data sharing and data management as well as reducing cooperation efficiency among road and enforcement agencies. In addition, due to the limited measurement capability of each system or device such as incomprehensive measurable traffic parameters, it has resulted in the total cost ineffective traffic data collection and monitoring, inefficient and ineffective speed and weight enforcement, and extremely difficult implementation of the actual regulation related to specific speed limit enforcement for heavy vehicle. The existing system or device for measuring dedicated traffic parameter can be integrated to provide an all-in-one integrated system for traffic data collection, monitoring and enforcement purposes. However, this method will definitely not be cost effective and time saving. This paper presents the development of an integrated quartz sensor based traffic data-collection and monitoring system, variable speed limit enforcement system and a weigh-in-motion based weight enforcement system. Results from field test show that the developed system complies with ASTM E1318-02 Type II and III Performance Requirements for all types of vehicle. This paper also provide some preliminary results to show the potential contribution of the system to transportation area.

Abstract

The purpose of this study is to extract important information from long distance travel data based on the Artificial Intelligence technique-rough set theory. Rough set theory can learn and refine decision rules or hidden facts from the incomplete observed data without the...Lead moreThe purpose of this study is to extract important information from long distance travel data based on the Artificial Intelligence technique-rough set theory. Rough set theory can learn and refine decision rules or hidden facts from the incomplete observed data without the constraints of statistical assumptions. In the study, we conducted a survey to collect the peoples’ most preferred travel mode choices for the given destination and people’s demographic information. We analyzed the observed data using rough set theory, calculated and discussed the approximation, core, reduct and rules of the data. The induced decision rules are expressed in natural language representing the relationships between personal demographic attributes and long distance travel mode choices, which can help policymakers in the decision making process. The results of validation test were very promising, which showed that the induced decision rules could represent the relationships between data with the accuracy of 74.59%.

Abstract

Car-following models describe the interaction between following vehicle's behaviour with respect to the preceding vehicle and are building blocks of various traffic simulation models. Car-following behaviour is well studied and analysed in the last fifty years. However, c...Lead moreCar-following models describe the interaction between following vehicle's behaviour with respect to the preceding vehicle and are building blocks of various traffic simulation models. Car-following behaviour is well studied and analysed in the last fifty years. However, conventional models express the following behaviour in mathematical form. In mixed traffic conditions, following behaviour depends on the vehicle-type and is much more complex to be predicted. A feed forward back propagation neural network model is proposed to predict this complex relationship. Performance of the neural network is studied using data collected from two urban arterials in Mumbai, India, and compared with the performance of widely used Gipps’ car-following model. Proposed neural network is able to predict the field behaviour closely. Trained neural network is embedded into a simulation program and the performance is analysed at macroscopic level.

Abstract

It is well known that a ramp metering is the most representative control strategy to solve an urban traffic problem such as congestion and an accident. This study is based on a pre-timed signal control used at the on-ramps of expressways in large cities, and it aims to si...Lead moreIt is well known that a ramp metering is the most representative control strategy to solve an urban traffic problem such as congestion and an accident. This study is based on a pre-timed signal control used at the on-ramps of expressways in large cities, and it aims to simulate and evaluate the effects on the traffic flow in accordance with a single- or multi-vehicle entry mode when ramp vehicles enter into the expressway mainline under the same metering rates. Results show that a multiple entry mode at the merge area affects much more on the mainline flow relatively than a single entry mode in terms of speeds, densities, delays, and capacity. Therefore, it is very important to select the entry mode whether a single vehicle enters or multiple vehicles enter during one signal cycle for the stability of traffic flow in expressway mainline.

Abstract

This paper presents an inter-vehicle geocasting and collision warning algorithm to accomplish multihop and collision avoidance in cooperative driving and data exchanged based on GPS positioning and dedicated short range communication (DSRC). The concept of cooperative col...Lead moreThis paper presents an inter-vehicle geocasting and collision warning algorithm to accomplish multihop and collision avoidance in cooperative driving and data exchanged based on GPS positioning and dedicated short range communication (DSRC). The concept of cooperative collision warning (CCW) system provides warning or situation awareness to drivers based on broadcast messaging the motions of neighboring vehicles obtained by one-way communication from other vehicles. The proposed system has the advantage of omnidirectional transmitting/receiving functions of DSRC module that provides 360-degree coverage in own surveillance region. The scenario of CCW system is mainly applied on team trips; moreover, the developed algorithm has the ability of message broadcasting and collision point estimation. This article concerns message geocasting and collision warning in inter-vehicle communication. Indeed, each CCW system can show other vehicles on the geographical map, and furthermore the developed inter-vehicle geocasting algorithm is the key point in remote surveillance, while some vehicles move away or turn into the roadway or intersection. The development system integrates DSRC and GPS with embedded system into CCW system. From the positioning data and vehicular information, this paper provides a conflict detection algorithm to do time separation and remote warning with error bubble consideration. The proposed system is carried out with theoretical application and hardware integration, and the result shows cooperative driving approach applicability using vector geocasting method. Some experiments are also scheduled about packet loss and latency by static tests, in which vehicles stop in fixed position away from 0 to 200 m.

Abstract

Pre-warning experimental system of driving fatigue has become a hotspot and difficulty of auto safety. Based on existing study and experimental equipment, the pre-warning experimental system of driving fatigue at the view of formation mechanism of driving fatigue was intr...Lead morePre-warning experimental system of driving fatigue has become a hotspot and difficulty of auto safety. Based on existing study and experimental equipment, the pre-warning experimental system of driving fatigue at the view of formation mechanism of driving fatigue was introduced in the paper. The system included data collection module, fatigue detection module, countermeasure generation module and countermeasure effectiveness validation module. The function of the system is to provide experiment platform for validating effectiveness of countermeasure against driving fatigue, to provide scientific and safe pre-warning signals for pre-warning system of driving fatigue. Firstly, the system detected and judge drivers’ fatigue state, once detected drivers fatigue, then the system gave pre-warning signals from sight, hearing ,feeling ,etc; Secondly, orthogonal experiment was used to design experiment program and study the effectiveness of various external stimulus to reduce driving fatigue; Finally ,the effective pre-warning signals against driving fatigue was proposed. It will provide theoretical basis for effective and safe pre-warning system of driving fatigue. The paper took the effectiveness of sound stimulus for example, introduced the method how to evaluate the effectiveness of pre-warning signals used the experiment system.

Abstract

This paper first analyzed the trait of travelers’ behavior of traffic mode choice based on the maximum utility theory and advanced the modified logit model under the condition of information supplying, which overcome the first class disadvantage named IIA character. The s...Lead moreThis paper first analyzed the trait of travelers’ behavior of traffic mode choice based on the maximum utility theory and advanced the modified logit model under the condition of information supplying, which overcome the first class disadvantage named IIA character. The second class of defect which could be stated that the probability of one mode had no connection with its utility was avoided through the modification and theoretic validation of utility influencing factors. The rationality of these two amendments was validated based on travelers’ choice behavior of traffic modes in Nanjing city. The fitting result shows that the forecasting precision of logit modified model is much higher than that of logit model, which reveals that the modification can explain travelers’ mental characters under the information supplying condition and it has an effective application meaning.

Abstract

Traffic congestion arises in metropolitans are common problems due to insufficient traffic facilities and major bottlenecks occur on main urban arterials, freeways and expressways. This research aims at proposing an optimal control model for corridor management, including...Lead moreTraffic congestion arises in metropolitans are common problems due to insufficient traffic facilities and major bottlenecks occur on main urban arterials, freeways and expressways. This research aims at proposing an optimal control model for corridor management, including freeway segments, on-ramps, off-ramps, and arterial streets, to relieve possible congestion. Traffic control strategies for corridor management include ramp metering, signal control, and route guidance. Based on the concept of the store-and-forward method, the problem is formulated as a linear optimal control model; and the objective function is to minimize total queue length for considered traffic corridors. Numerical experiments are conducted to illustrate the proposed models.

Abstract

In this paper, we propose the pedestrian detection method using the local pedestrian classifier. Because the pedestrian detection is considered a problem to distinguish a pedestrian from the background around the observation point, the proposed method divides the route al...Lead moreIn this paper, we propose the pedestrian detection method using the local pedestrian classifier. Because the pedestrian detection is considered a problem to distinguish a pedestrian from the background around the observation point, the proposed method divides the route along which a vehicle runs into several local sections and creates the local pedestrian classifier by learning background images collected in each local section as non-pedestrian samples. Each local pedestrian classifier is tied to position data obtained from GPS system. When the vehicle runs, the self-location of the vehicle is measured by the GPS system and the local pedestrian classifier in the local section closest to the vehicle is selected to discriminate whether or not pedestrians exist. The coarse HOG feature named 2HOG, the fine HOG features named 18HOG and the Support Vector Machine(SVM) are used to create local pedestrian classifiers. The 2HOG feature consists of 2-bin histogram representing horizontal direction and vertical direction. The 18HOG feature consists of 18-bin histogram representing eighteen directions between 0 and 360 degree. The cascade connection of the 2HOG classifier and the 18HOG classifier realizes efficient pedestrian detector because the 2HOG feature can detect pedestrian candidates quickly and the 18HOG feature can examine the pedestrian contours in detail. Experimental results for real road scenes show the effectiveness of the proposed method.

Abstract

This paper proposes a method of lane marker detection for platooning and automatic driving technology. The proposed method recognizes a lane markers in images captured from a front camera and downward side cameras. First, candidate points of lane markers are extracted by ...Lead moreThis paper proposes a method of lane marker detection for platooning and automatic driving technology. The proposed method recognizes a lane markers in images captured from a front camera and downward side cameras. First, candidate points of lane markers are extracted by edge pair detection. Then, straight lines representing lane marker are detected by applying Hough transform to these candidate points. In a front camera, a yaw angle of a vehicle is calculated by estimating an intersection point of straight lines. And, in a downward side camera, a lateral position is estimated from a position of a straight line in an image. These methods were implemented in the image processing hardware whose CPU satisfies on-vehicle specifications. Experimental results show the effectiveness of the proposed method and a lane detection device. In experiments conducted at the test course, the vehicle ran at 80 km/h along the straight lane automatically.

Abstract

Since precipitation causes negative impacts on the traffic congestion, there have been various studies for modeling the relationship between precipitation and its impact on traffic flow. Due to existing data limitation issue, however, none of the previous studies has cond...Lead moreSince precipitation causes negative impacts on the traffic congestion, there have been various studies for modeling the relationship between precipitation and its impact on traffic flow. Due to existing data limitation issue, however, none of the previous studies has conducted for estimating total delay caused by precipitation with respect to each weather event. The objective of this study is to estimate non-recurrent traffic congestion on freeways caused by precipitation. To accomplish this objective, archived weather and traffic data for the year 2008 collected from the Korean Freeway Systems were combined to estimate and analyze non-recurrent congestion impacts caused by precipitation. As a result, non-recurrent traffic congestion was about 16 million vehicle-hours for rainfall and 2 million vehicle-hours for snowfall in 2008. In addition, the simple analyses were performed that describe average non-recurrent traffic congestion as a function of the cumulative precipitation and unit distance and a function of the time period of precipitation. Although the precipitation events might not be handled by human factors, these results will beneficially help in making strategic plans for mitigating traffic congestion due to precipitation.

Abstract

Since accident duration is defined as the time difference between when accident occurred and when the response vehicles depart the accident scene, understanding the relation between characteristics of an accident and its duration is crucial for the efficient response of a...Lead moreSince accident duration is defined as the time difference between when accident occurred and when the response vehicles depart the accident scene, understanding the relation between characteristics of an accident and its duration is crucial for the efficient response of accidents and the reduction of traffic congestion caused by accidents. However, the previous studies have been focused on modeling accident duration rather than its observation. If considering a fallacy due to the difficulty of observation on accident scene from its occurrence to clearance, the collection of the duration data is crucial as well. Thus the first objective of this study is to develop an analytical method to observe duration caused by freeway traffic accident based on the archived travel speed profile. Additionally, a statistical analysis for accident duration is presented based on the observation result from the case study involving approximately 6,200 accidents that occurred on freeways in Orange County, California, in 2001. The analysis results were consistent with the theoretical expectation and the prior studies. Overall, the presented method will contribute to less observation error and burden for accident duration.

Abstract

Travel time prediction for urban networks is an important issue in Advanced Traveler Information Systems (ATIS) since drivers can make individual decisions, choose the shortest route, avoid congestions and improve network efficiency based on the predicted travel time info...Lead moreTravel time prediction for urban networks is an important issue in Advanced Traveler Information Systems (ATIS) since drivers can make individual decisions, choose the shortest route, avoid congestions and improve network efficiency based on the predicted travel time information. In this research, two algorithms are proposed to estimate and predict travel time for urban networks, the simulation-based and time-series models. The simulation-based model, DynaTAIWAN, designed and developed for mixed traffic flows, is adopted to simulate the traffic flow patterns. The Autoregressive Integrated Moving Average (ARIMA) model, calibrated with vehicle detector (VD) data, is integrated with signal delay to predict travel time for arterial streets. In the numerical analysis, an arterial street in Kaohsiung city in Taiwan is conducted to illustrate these two models. The empirical and historical data are used to predict and analyze travel time, including: travel time data from survey and historical speed data from vehicle detector (VD).

Abstract

Traffic Light Assistance (TLA) system based on Infrastructure-Vehicle Communication (I2V) provides a new effective approach for the solutions of safety and traffic problems at intersections. A TLA algorithm is purposed which combines two main functions of passing support ...Lead moreTraffic Light Assistance (TLA) system based on Infrastructure-Vehicle Communication (I2V) provides a new effective approach for the solutions of safety and traffic problems at intersections. A TLA algorithm is purposed which combines two main functions of passing support and traffic signal violation warning. Passing support assists the driver to pass the intersection without stop by providing speed recommendation, whereas traffic signal violation warning warns the driver against potential traffic signal violation. A hybrid TLA algorithm is found by matching the two main functions and an additional function of rear-end collision warning with 19 typical intersection driving scenarios. A simulation model is built to verify the algorithm and access its impacts on intersection driving. The results of simulation show that the purposed hybrid TLA algorithm can simultaneously improve driving safety, comfort and smoothness at intersections.

Abstract

Among many ITS subsystems, the ATIS (Advanced Traffic Information System) is one of the most widely used subsystems. Traffic information devices such as VMS (Variable Message Signs), the internet, radio broadcasts and cellular phones are commonly used to provide traffic i...Lead moreAmong many ITS subsystems, the ATIS (Advanced Traffic Information System) is one of the most widely used subsystems. Traffic information devices such as VMS (Variable Message Signs), the internet, radio broadcasts and cellular phones are commonly used to provide traffic information to road users. The main purpose of providing traffic information to road users is to help them find or decide upon the best driving solution considering the current traffic conditions. Using the given traffic information, road users are able to find the minimum time paths to their destinations. Road users also receive psychological benefits in addition to the real time savings by receiving traffic information. In this study, we develop user satisfaction models to evaluate various traffic information devices based upon the road users’ perceived benefits from different traffic information. First, we evaluate the benefits of various traffic information devices. A number of factors are investigated to assess the impact ot the benefits. Based upon these factors, user satisfaction models are constructed. The results show that the factors which influence the effectiveness of traffic information devices are different from the devices. Among those factors, the effectiveness of the given traffic information which influences the selection of routes proved to be the most important factor.

Abstract

The primary focus of this paper is to present an integrated control model for urban freeway using Model-based Predictive Control (MPC) framework, where the effect of traffic information on driver’s dynamic enroute diversion behavior has been explicitly modeled by using an...Lead moreThe primary focus of this paper is to present an integrated control model for urban freeway using Model-based Predictive Control (MPC) framework, where the effect of traffic information on driver’s dynamic enroute diversion behavior has been explicitly modeled by using an Adaptive Constrained Kalman Filtering (ACKF) technique, so as to capture the effect of traffic information on driver’s enroute diversion behavior in real time based on-line traffic surveillance data. The integrated control task is formulated as a dynamic, nonlinear, discrete time optimal control problem with constrained control variables. The network traffic process is modeled by a macroscopic network traffic flow model, which is deterministic, discrete time, discrete space, and macroscopic and suitable for model based traffic control. Feedback control is realized by solving the optimization problem for each control interval over a sufficiently long future time horizon. Simulation results for a case study show that the proposed integrated MPC traffic control can account for the time-dependent traffic characteristics and driver’s actual enroute diversion behavior and guarantee a high degree of control benefits.

Abstract

115 VMSs have been installed on urban expressway in Shanghai since 2003 and traffic state information on expressway was posted to assist drivers in making more informed decisions on route selection. In Traffic Guidance Piloting Project founded in 2008, seven sets of VMSs ...Lead more115 VMSs have been installed on urban expressway in Shanghai since 2003 and traffic state information on expressway was posted to assist drivers in making more informed decisions on route selection. In Traffic Guidance Piloting Project founded in 2008, seven sets of VMSs as well as vehicle license plate recognizers and loop detectors are installed on expressway and some surface-streets in the demonstration area. In this paper, data from loop detectors since 2003 and vehicle license plate recognizers since 2008 are used to evaluate the impact of VMS messages on drivers’ en-route diversion behavior. We found an aggregated en-route diversion model based on real-detected data, and find that drivers are more sensitive to travel time information than traffic state information; their en-route behavior will be more rational if information of alternate routes is provided and neighboring VMSs are coordinated. These results are accordant with that of SP survey. In real road network, from the sensitivity analysis of drivers’ en-route behavior, we find that the content of information, peak hours or other periods, off-ramp condition, visibility of downstream congestion have significant influence on en-route behavior, which can explain the obvious difference between the result from en-route model based on SP survey and the real traffic system.

Abstract

In this paper, we propose a design methodology of GPS-based floating vehicle data (GFVD) system by using historical data, real-time GPS floating vehicle data and data fusion technology to provide instant traffic information collection for Kaohsiung city. There are more th...Lead moreIn this paper, we propose a design methodology of GPS-based floating vehicle data (GFVD) system by using historical data, real-time GPS floating vehicle data and data fusion technology to provide instant traffic information collection for Kaohsiung city. There are more than 500 taxies and 850 fleet dispatching vehicles served as probing cars and join the fleet management project to provide instant traffic information. We utilize the GPS data in our GFVD system to calculate traffic information instead of vehicle detection (VD) deployed in Kaohsiung. From the historical data and fusion algorithm, the completed general road sections are 8898 roads and with 764 problems road sections, which come from 453 with wrong section of roads, without correct TL road names, 15 wrong doing roads. The correct percentage of GFVD is 91.4 % in Kaohsiung City. Thus, GFVD system can provide a cheaper and reliable data with instant traffic information for any wait-for-test road. It can compensate for the shortage of infrastructure construction in a big city.

Abstract

The purpose of this research is to simulate realistic traffic flow with adequate high reproduction rate as to be able to evaluate ITS technologies and policies with various criteria. To do this, a method to calibrate driver model parameters with consideration of vehicle g...Lead moreThe purpose of this research is to simulate realistic traffic flow with adequate high reproduction rate as to be able to evaluate ITS technologies and policies with various criteria. To do this, a method to calibrate driver model parameters with consideration of vehicle generation distribution is proposed implementing the travel time estimation method, adding to calibration of traffic volume and average speeds observed by traffic detectors. In the previous study with our microscopic traffic simulator KAKUMO, the parameters were calibrated and evaluated macroscopically by the traffic volume and average speed in each lane obtained from traffic detector data. Using the travel time estimation method, the reproduction of simulated traffic is considered to be evaluated in multiple viewpoints with travel time along a section calculated from detector data at certain points on the assumption with an appropriate vehicle generated distribution.

Abstract

The most common taxi operation service is conventional cruising randomly around the city to search for passengers. Because of de-regulation policy, over-supply and high vacancy rate have caused inefficient operation and low performance in taxi market. Therefore, this pape...Lead moreThe most common taxi operation service is conventional cruising randomly around the city to search for passengers. Because of de-regulation policy, over-supply and high vacancy rate have caused inefficient operation and low performance in taxi market. Therefore, this paper aims to create an innovative operation service and evaluate the environmental benefits compared with the conventional cruising services. It is shown that innovative dispatching service can have a 4.35 billion USD of energy saving per year in Taipei Metropolitan. Numerical results have also shown other external effects of innovative dispatching service, including less air pollution, traffic accidents and traffic congestion.

Abstract

Traffic simulation studies can be used to evaluate and optimize policy choices that affect adaptive traffic control (ATC) systems. Appropriate modeling accuracy of an ATC system is required to ensure that study outcomes are defensible. The Sydney Coordinated Adaptive Traf...Lead moreTraffic simulation studies can be used to evaluate and optimize policy choices that affect adaptive traffic control (ATC) systems. Appropriate modeling accuracy of an ATC system is required to ensure that study outcomes are defensible. The Sydney Coordinated Adaptive Traffic System (SCATS) provides a responding facility ? known as SCATSIM. When using ATC systems within simulation there is a considerable challenge of the appropriate design of the study; we refer to this design as the ‘modeling experimental design’. We present details of the modeling experimental design of a comprehensive SCATSIM study that considered the intricacies of SCATS operation. This example provides modelers and stakeholders with a practical insight into: the rigorous experimental design of a SCATSIM study, and the simulation of ATC systems, generally.

Abstract

In this research, we developed a driver’s drowsiness monitoring system consisting of a gas pedal force sensor, a steering wheel grip force sensor and a headrest sensor. It was hypothesized that a vigilant driver would respond more frequently and quicker to changes in road...Lead moreIn this research, we developed a driver’s drowsiness monitoring system consisting of a gas pedal force sensor, a steering wheel grip force sensor and a headrest sensor. It was hypothesized that a vigilant driver would respond more frequently and quicker to changes in road situations than a less vigilant driver. In other words, a drowsy driver shows monotonous response to a gas pedal and a steering wheel during driving. A driving simulator was used to implement the driver monitoring system. Pressure sensors were installed on a gas pedal, a steering wheel and a headrest in a driving simulator cabin. Pedal and grip force were measured to determine their jerk and then the jerk was compared with the moving averages. A driver status determination algorithm was developed based on a series of critical values for the difference between the jerk and the moving average. Driving simulator experiment was conducted to evaluate the monitoring system.

Abstract

This paper presents an investigation result that is made to determine dominant influencing factors and their specific impacts on motorist detour makings for VMS placements on freeway. This was necessary to provide engineers with a tool for evaluating VMS cost-effectivenes...Lead moreThis paper presents an investigation result that is made to determine dominant influencing factors and their specific impacts on motorist detour makings for VMS placements on freeway. This was necessary to provide engineers with a tool for evaluating VMS cost-effectiveness and its feasibility. This research found that additional travel time delay is the most influential variable in determining motorist detour makings. Also, more than 70% freeway motorists recognize the adjacent national roads as their detour routes, and both AADT and the distance to the entering point of the detour path are found to be dominating factors. Motorist detour ratio model by VMS placement on freeway is developed based on these observations. In model validation, actual detour making behaviors observed by the Revealed Preference values show some discrepancies, but these values are minimal. These research findings should be useful for cost-effective placements of VMS on freeway, and it is recommended that the results be tested by practicing engineers in the ITS sector.

Abstract

13,905-km-long national highway network connecting the entire Korea, especially rural areas, has 12,111 intersections. Over 54 percent of the 12,111 intersections shows the typical geometry of T-intersections of a major street with a minor crossroad where the traffic is s...Lead more13,905-km-long national highway network connecting the entire Korea, especially rural areas, has 12,111 intersections. Over 54 percent of the 12,111 intersections shows the typical geometry of T-intersections of a major street with a minor crossroad where the traffic is small or few. The majority of the intersections generally operate under pretimed signal control, which has limitations in reflecting traffic fluctuations common in national highways in rural areas. Therefore, this research effort aims at developing a signal control strategy based on semi-actuated signal control and a video detection technology which is able to improve the operational efficiency of traffic signal control on such T-intersections. In the evaluation using a calibrated CORSIM network with various volume scenarios, the proposed left-turn actuated control with the volume-density control feature (LVSC) was more efficient so that it can reduce the control delay by 36.4 % and 11.8 % compared with the pre-timed control in use and a typical left-turn actuated control.

Abstract

The city of Suwon has introduced various ITS services between 2001 and 2006 as an alternative to solve chronic traffic congestion in the city. After the successful implementation, there was an effort to evaluate effects from the ITS deployments. However, the effect caused...Lead moreThe city of Suwon has introduced various ITS services between 2001 and 2006 as an alternative to solve chronic traffic congestion in the city. After the successful implementation, there was an effort to evaluate effects from the ITS deployments. However, the effect caused from ITS implementations may be realized in many different ways. Therefore, this research effort aims at evaluate the effectiveness of ITS implements through the first and second phases in Suwon using reasonable and scientific methods and processes. Based on results from the analysis of variables presenting traffic condition, surveys and economic evaluation, the implementation of ITS services in Suwon can be concluded successful. However, there is a need for efforts to maintain the effects produced from the successful introduction of ITS services through continuous upgrading the systems, intensive maintenance, and efficient operations.

Abstract

Travel time is considered to be more useful to users than other travel related information such as speed. It can be mainly estimated by point detection systems or spatial detection systems. This paper investigated the deficiency of these detection systems when they are us...Lead moreTravel time is considered to be more useful to users than other travel related information such as speed. It can be mainly estimated by point detection systems or spatial detection systems. This paper investigated the deficiency of these detection systems when they are used solely in estimating travel times. To resolve it, a fusion algorithm is suggested, which is simultaneously utilizing the data from both of point detection systems and spatial detection systems. The proposed fusion algorithm is established based on the traffic flow model and the k-nearest neighborhood(k-NN) model. The study precisely defined so called the time-lag issue originated from the spatial detection systems. To overcome it, the travel time variation originated from the traffic states variation is analyzed using the fused data from point detection systems and spatial detection systems. Considering the travel time variation, the travel time is renewed which is collected from spatial detection systems. The results show that the proposed algorithm could predict satisfactorily the travel time with the MAPE(Mean Absolute Percentage Errors).

Abstract

VANETs (vehicular ad hoc networks) is a promising technology to facilitating road safety and traffic management for drivers and passengers. Recently, it is considered that security and protection of private user information are a prerequisite for the deployment of the tec...Lead moreVANETs (vehicular ad hoc networks) is a promising technology to facilitating road safety and traffic management for drivers and passengers. Recently, it is considered that security and protection of private user information are a prerequisite for the deployment of the technology. This is because the benefits of VANETs could be disrupted or disabled, even by relatively incapable adversaries. Thus one of the ultimate goals in the design of such network is to protect various malicious abuses and security attacks. In this paper, we first address secure vehicular communication scenario, which is one of the VANETs structures. We then address four security modules which are the components of this scenario. And implementation and security analysis what we perform simulation are introduced.

Abstract

This paper presents the impact of information on travel mode choice by proving two hypotheses. A new definition named the factor tie was firstly brought forth and then based on which, two assumptions were given. One assumed that information is the factor tie in the proces...Lead moreThis paper presents the impact of information on travel mode choice by proving two hypotheses. A new definition named the factor tie was firstly brought forth and then based on which, two assumptions were given. One assumed that information is the factor tie in the process of travel mode choice analysis while the other supposed that the two-layer relationship model (TLRM), an academic model advanced in this paper, explained the first hypothesis. Based on the two assumptions, data collected from Nanjing city in China was studied and the result was consistent with the two hypotheses. Information plays a role of factor tie in the process of travel mode choice behavior. At the upper layer, it updates alterable variables. Respondents are apt to reelect travel modes according to the change of alterable variables. While at the lower layer, it is interactive with personal attributes. By proving these two assumptions, it concludes that alterable information will lead to more travel mode reelection.

Abstract

This paper presents a simple method for short term prediction of freeway traffic collisions based on the traffic flow theory. A model is proposed to estimate the probability of traffic collision occurrence, which is achieved by analyzing the real time traffic flow data fr...Lead moreThis paper presents a simple method for short term prediction of freeway traffic collisions based on the traffic flow theory. A model is proposed to estimate the probability of traffic collision occurrence, which is achieved by analyzing the real time traffic flow data from two neighboring traffic detectors. The proposed model indicates that the probability of traffic collision occurrence is positively related to two terms: a) the ratio of the upstream traffic density to the steady traffic density, and b) the speed difference between the upstream traffic and the downstream traffic. The empirical data from Dutch motorways is used to calibrate the model. The proposed model only needs fundamental traffic data to estimate traffic collision occurrence probability. It appears promising for the application in the on-line traffic incidents predictions in the future.

Abstract

School bus is a convenient and efficient transportation mode. Useful travel time information can help commuters make good choice of departure time as well as other travel decisions. However, providing real time travel time information for an en route school bus is a diffi...Lead moreSchool bus is a convenient and efficient transportation mode. Useful travel time information can help commuters make good choice of departure time as well as other travel decisions. However, providing real time travel time information for an en route school bus is a difficult task as travel time information is uncertain due to external interferences. Hence no consistent tendency can be easily observed. In this research, we develop a novel approach that combines grey theory (for travel time prediction) and empirical decomposition method (for data analysis) to predict the arrival time at each stop along the school bus route. The experiments showed that our prediction approach employing real time data collected from geographic positioning system as well as historic data outperforms the approach involving only historic data. In addition, sensitivity analysis is performed for different number of stops used in the prediction process.

Abstract

For the development of crash cushion, it relies on the full-scaled vehicle crash test in general. However, collision behavior of crash cushion can usually be observed only for less than 0.4 second. The very short duration time make it hard or impossible to analyze the cra...Lead moreFor the development of crash cushion, it relies on the full-scaled vehicle crash test in general. However, collision behavior of crash cushion can usually be observed only for less than 0.4 second. The very short duration time make it hard or impossible to analyze the crash numerically. To make it easy and possible, we are using various kind of Data Acquisition Unit. And all data from the DAU are processed to generate meaningful values to evaluate the tested article. For example, occupant safety indices which are for the evaluation of occupant risk are calculated using the data acquired from the DAU. And collision behaviors of test vehicle and crash cushion are traced by high-speed cameras during the test. But the range of use of the high-speed camera is very limited and it is not utilized in calculating the occupant safety indices. Since emerging technology of high-speed camera even make it possible to analyze the energy dissipating mechanism of crash cushion, the data from the cameras can be involved in the evaluation process of occupant risk. This paper is to ascertain the appicability of the data from high-speed camera in a full-scaled vehicle crash test. Our goal is to find the probability of substitution of the DAU with those cameras. For that we compared two data sets of crash results 18 times which had been achieved in our test field on various types of crash cushions for last few years.

Abstract

This research is for analyzing uses of vehicle navigations and for providing appropriate information for drivers. Since recommended route from navigation is based on the shortest path, there is a difference between actual driving route and the route from navigation system...Lead moreThis research is for analyzing uses of vehicle navigations and for providing appropriate information for drivers. Since recommended route from navigation is based on the shortest path, there is a difference between actual driving route and the route from navigation system based on the characteristics of drivers and the purpose of driving. Throughout analyzing log historical data, compliance rate showed about 70%. When we compared the number of route changes per km, the study shows that the number of route changes of week days is 0.7~3 times higher than weekends’. We analyzed the main cause of route changes by deriving structure equation. The data shows that distance is 45% in the direct effect criteria, and experience is 36% higher than usual. Furthermore, indirect effect from route is higher when the route is much simpler and when Congestion is heavier than usual. It is necessary to consider making simpler route in heavy traffic areas and making better information of navigation by considering the route which many people actually drive frequently. This study will make it to be available to understand information of car navigation systems by theoretical research which is considering traffic characteristics.