Motohiro Fujita

Predicting driver’s lane-changing decisions using a neural network model

Abstract Lane changing has a significant impact on traffic flow characteristics and potentially reduces traffic safety. However, literature relating to lane changing is not comprehensive, largely owing to the inherent complexity of lane changing and a lack of large-scale data to analyze such behavior. In an effort to cope with these obstacles, this study adopts a neural network (NN) model to capture the complexity of lane changing, and large-scale trajectory data are employed for model estimation and validation. For comparison purposes, a multinomial logit (MNL) model that was frequently accepted as a framework for lane changing in previous studies is also built. Although for non-lane-changing samples, both models perform well in model estimation and validation processes, for lane-changing samples, there are significant differences in their performance. The NN model is able to correctly predict 94.58% of left lane-changing samples and 73.33% of right lane-changing samples in the model estimation process, whereas the percentage correctly predicted by the MNL model is only 13.25% and 3.33%, respectively. While the accuracy of both models noticeably drops in the model validation process, prediction results in the NN model are still acceptable. Finally, the impact of heavy vehicles on driver’s lane-changing decisions is quantitatively evaluated using the sensitivity analysis of the proposed NN model.

Modelling a vehicle's speed fluctuation with a cellular automata model

As is well known, a vehicle’s speed fluctuation has significant impact on traffic capacity, road safety, fuel consumption, and exhaust gas emission. Considerable microscopic traffic models have been developed in past decades, while the ability of describing the realistic speed fluctuation has rarely been examined. With the data from real traffic, in this study we investigate the performance of two typical car-following models on modelling speed fluctuation. Our findings indicate that neither of them can mimic a realistic speed fluctuation with high accuracy. In addition, it is found that the model with the minimum speed error does not necessarily mean it can describe speed fluctuation most realistically. To simulate this phenomenon more accurately, by introducing a reasonable duration of stable speed, we propose one kind of cellular automata model. Simulation results show that the model depicts a vehicle’s speed fluctuation with higher fidelity, relative to two typical models.

Integrating Stochastic Failure of Road Network and Road Recovery Strategy into Planning of Goods Distribution After a Large-Scale Earthquake

Disaster relief operations are complex and can benefit greatly from a high level of preparedness. One of the main sources of complexity in disaster operations is uncertainty. An analysis of a disaster relief operation in Aichi Prefecture, Japan, preparing for the periodic Tokai–Tonankai earthquake is presented. In this study, the possible degradation of the road network is considered by including a stochastic element to represent the possibility of link failure dependent on earthquake intensity in each subregion. Also, a strategy to fix the roads is integrated into the analysis to evaluate its impact on the disaster logistics operations. The analysis is performed with the current road network of Aichi Prefecture. The results suggest the best preparation of resources and identify vulnerable destinations that are most likely to be cut off by the disaster. Also analyzed is a relocation of hubs that can reduce the total response time and take into account the possibility that some links will be destroyed. This analysis is important to help planners to evaluate their strategies, to identify vulnerable locations, and to be able to prepare in advance the best methods to deal with the uncertainty of road failure.

A Car-Following Model Based on Discrete Choice Theory

In the car-following model proposed in this paper, the acceleration and deceleration are incorporated into seven alternatives, and alternative-specific parameters are adopted to capture different attractiveness of each alternative. To avoid interferences from lane-changing behavior, the model is estimated and validated by trajectory data in the high occupancy vehicle (HOV) lane. From the estimation results, most estimated parameters are significantly different from zero at 95% confidence level, which indicates that the chosen variables and the classification of alternatives are effective and reasonable. In addition, according to the proposed model, the changing tendencies of choosing different alternatives at different speeds and gaps are completely consistent with common driving behaviors. The validation results show that, for the chosen alternatives, the percentages of predicted probabilities that are larger than 1/7 (the average probability of choosing each alternative) are more than 87% in the used data sets. From the shares of observations point of view, there are only minor differences between the observed and predicted results. Finally, this model is applied to simulate the 30-min single lane traffic conditions. Through simulation results, it can be found that the proposed model can accurately represent the real traffic conditions on the macroscopic level, while, on the microscopic level, there are some defects needed to be further improved.

Stochastic modeling of road system performance during multihazard events: Flash floods and earthquakes

AbstractTransport resilience is an important area of research in the global effort to adapt to climate change. This paper introduces and applies a stochastic modeling methodology to assess the impa...

Performance Evaluation of a Crossing-Assistance System for Visually Disabled Persons at Intersections Considering Actual Road Structure

This paper evaluates, through outdoor experiments and a questionnaire, the performance of a crossing-assistance system for visually disabled persons at signalized intersections by utilizing pedestrian traffic signals in concert with visible light communication (VLC) technology. From the results of the survey, it is found that the guidance given by the VLC system enables the visually disabled to cross safely at signalized intersections, under actual road conditions. The applicability of the system is highly evaluated.

Time Coefficient Estimation for Hourly Origin-Destination Demand from Observed Link Flow Based on Semidynamic Traffic Assignment

Day-long origin-destination (OD) demand estimation for transportation forecasting is advantageous in terms of accuracy and reliability because it is not affected by hourly variations in the OD distribution. In this paper, we propose a method to estimate the time coefficient of day-long OD demand to estimate hourly OD demand and to predict hourly traffic for urban transportation planning of a large-scale road network that lacks discrete-time rich traffic data. The model proposed estimates the time coefficients from observed link flows given a proven day-long OD demand based on a bilevel formulation of the generalized least square and semidynamic traffic assignment (OD-modification approach). The OD-modification approach is formulated as a static user-equilibrium assignment with elastic demand, based on the residual demand at the end of each period. Our model does not require setting many parameters regarding the OD demand matrices and the discrete-time dynamic traffic assignments. Applying the model to large-scale road network demonstrates that it efficiently improves estimation accuracy because the 24-hour time coefficients of survey data are slightly biased and may be modified properly. In addition, the methods that partially relax the assumption of OD-modification approach and transform the estimated demand into demand based on departure time are examined.

Analysis on Effective Information and Travel Activity Returning Home by Car under Downpour

The way of providing information for reducing disaster damage has been more important in Japan. By providing appropriate information to citizens under downpour, citizens can understand disaster situation and behave appropriately. Therefore, this paper focuses on the use of privately owned cars under downpour which is often brought up the deteriorating factor of traffic jam under downpour, and examines the way of providing traffic & weather information for reducing car use under downpour through the development of behavior model of returning home and other analyses. As a result, it found that the effect factor to driver consciousness of returning home under downpour.

Study On Road Traffic Conditions Under Concentrated Repair Work For Urban Expressway

In this paper we analyze Road traffic conditions and Traffic actions under concentrated repair Work, based on some investigations by Nagoya Expressway Company. We apply Time-of-Day traffic equilibrium assignment model (TUE) to Nagoya metropolitan road network considering concentrated repair work and the results show good accuracy. Moreover if TUE is applied considering “Change Traffic Actions”, the results show better accuracy than without considering. We consider traffic policys such as Guidance to Detour to manage future concentrated repair works, and we analyze the effect of traffic policys.

EVALUATION OF CONGESTION ON URBAN STREETS WITH SIGNALIZED INTERSECTIONS

We propose a new definition of congestion for urban streets with signalized intersections on the basis of drivers consciousness. The new one is defined by both average speed including signalstops and travel time in congested road area. It is shown that the new definition fit the Blochs law which is famous for the consciousness study of psychology. And it is shown that if we use the new definition in order to decide congested road area on urban streets, we can obtain the result which really fit drivers consciousness.