Neural networks have been extensively applied to short-term traffic prediction in the past years. This study proposes a novel architecture of neural networks, Long Short-Term Neural Network (LSTM NN), to capture nonlinear traffic dynamic in an effective manner. The LSTM NN can overcome the issue of back-propagated error decay through memory blocks, and thus exhibits the superior capability for time series prediction with long temporal dependency. In addition, the LSTM NN can automatically determine the optimal time lags. To validate the effectiveness of LSTM NN, travel speed data from traffic microwave detectors in Beijing are used for model training and testing. A comparison with different topologies of dynamic neural networks as well as other prevailing parametric and nonparametric algorithms suggests that LSTM NN can achieve the best prediction performance in terms of both accuracy and stability.

Long short-term memory neural network for traffic speed prediction using remote microwave sensor data

This article presents the theoretical basis for modeling univariate traffic condition data streams as seasonal autoregressive integrated moving average processes. This foundation rests on the Wold decomposition theorem and on the assertion that a one-week lagged first seasonal difference applied to discrete interval traffic condition data will yield a weakly stationary transformation. Moreover, empirical results using actual intelligent transportation system data are presented and found to be consistent with the theoretical hypothesis. Conclusions are given on the implications of these assertions and findings relative to ongoing intelligent transportation systems research, deployment, and operations.

/pdf/modeling-and-forecasting-vehicular-traffic-flow-as-a-4l9jx8epis.pdf

Modeling and Forecasting Vehicular Traffic Flow as a Seasonal ARIMA Process: Theoretical Basis and Empirical Results

For the last two decades, intelligent transportation systems (ITS) have emerged as an efficient way of improving the performance of transportation systems, enhancing travel security, and providing more choices to travelers. A significant change in ITS in recent years is that much more data are collected from a variety of sources and can be processed into various forms for different stakeholders. The availability of a large amount of data can potentially lead to a revolution in ITS development, changing an ITS from a conventional technology-driven system into a more powerful multifunctional data-driven intelligent transportation system (D2ITS) : a system that is vision, multisource, and learning algorithm driven to optimize its performance. Furthermore, D2ITS is trending to become a privacy-aware people-centric more intelligent system. In this paper, we provide a survey on the development of D2ITS, discussing the functionality of its key components and some deployment issues associated with D2ITS Future research directions for the development of D2ITS is also presented.

Data-Driven Intelligent Transportation Systems: A Survey

Traffic congestion in urban road and freeway networks leads to a strong degradation of the network infrastructure and accordingly reduced throughput, which can be countered via suitable control measures and strategies. After illustrating the main reasons for infrastructure deterioration due to traffic congestion, a comprehensive overview of proposed and implemented control strategies is provided for three areas: urban road networks, freeway networks, and route guidance. Selected application results, obtained from either simulation studies or field implementations, are briefly outlined to illustrate the impact of various control actions and strategies. The paper concludes with a brief discussion of future needs in this important technical area.

/pdf/review-of-road-traffic-control-strategies-2pzcx3izzi.pdf

Review of road traffic control strategies

Since the early 1980s, short-term traffic forecasting has been an integral part of most Intelligent Transportation Systems (ITS) research and applications; most effort has gone into developing methodologies that can be used to model traffic characteristics and produce anticipated traffic conditions. Existing literature is voluminous, and has largely used single point data from motorways and has employed univariate mathematical models to predict traffic volumes or travel times. Recent developments in technology and the widespread use of powerful computers and mathematical models allow researchers an unprecedented opportunity to expand horizons and direct work in 10 challenging, yet relatively under researched, directions. It is these existing challenges that we review in this paper and offer suggestions for future work.

Short-term traffic forecasting: Where we are and where we’re going

Road transportations still play a dominant role in goods delivery, and driving behaviours significantly affect the fuel economy of heavy-duty trucks. Plenty of fossil fuel is wasted as a result of ...

Analysis of driving behaviours of truck drivers using motorway tests

Indirect Reinforcement Learning for Incident-responsive Ramp Control

In this paper, the authors present a method of estimating the probability distribution of incident-free data by using Principal Component Analysis. This technique uses novelty detection by measuring the input's distance to the centroid of the normal data. This technique is applicable to incident detection in dynamic traffic monitoring systems.

Motorway incident detection using principal component analysis

Traffic noise has been known to severely affect human population. The valuation of traffic noise pose a significant challenge in choice experiments as respondents have little understanding of the physical measure of noise and its associated perception. As a result, several techniques have been developed that represent noise using different methods, either based on the level of noise exposure or the respondent’s level of noise annoyance. This study examines the effect of different methods of attribute representation on respondents’ attribute understanding and valuation. The study is focussed on residential choice and residential view and sunlight are important attributes that are examined along with traffic noise. The study demonstrates that the methods of attribute representation have an important effect on respondents’ understanding of the attributes as well as in the subsequent valuation. It was found that attribute such as view is better represented using the location representation while noise is better represented using the linguistic method. Moreover, the method of attribute modelling also plays a significant role in the analysis as certain data input techniques are more suitable for some representation methods.

/pdf/the-effect-of-attribute-representation-methods-on-noise-4vyh4euga2.pdf

Haibo Chen

Papers

Analysis of driving behaviours of truck drivers using motorway tests

Indirect Reinforcement Learning for Incident-responsive Ramp Control

Motorway incident detection using principal component analysis

The effect of attribute representation methods on noise valuation: A choice experiment study

Variations of significant contribution regions of NOx and PN emissions for passenger cars in the real-world driving