Real-time Traffic Prediction Research Articles

Spatial–Temporal Deep Tensor Neural Networks for Large-Scale Urban Network Speed Prediction

Real-time traffic speed prediction is an essential component of intelligent transportation systems applications on large-scale urban networks, e.g., proactive traffic management, advanced information provision, and prompt incident response. The family of traffic prediction models (e.g., convolutional neural networks) based on multi-detector speed diagrams in the time-space plane has been one of the most frequently used approaches for individual roads and the entire network. However, the predefined stacking sequence of traffic detectors along the spatial dimension of the speed diagram has a significant influence on the prediction performance, which makes network-wide speed prediction more challenging. To tackle the above challenge and better capture complicated traffic dynamics, we propose a novel speed prediction approach, named spatial-temporal deep tensor neural networks (ST-DTNN), for a large-scale urban network with mixed road types. Spatial and temporal dependencies of different road segments are simultaneously taken into account to improve the network-wide prediction accuracy. A scalable deep tensor is constructed for the ST-DTNN to eliminate the potentially negative impact caused by the manually stacking sequence of speed time series collected at different locations. Multi-step ahead traffic speeds can be simultaneously predicted based on probe data for a real-world large-scale urban network with hundreds of detectors installed on freeways, highways, and major/minor arterials. The results demonstrate the capability and effectiveness of the proposed ST-DTNN approach. Compared with the benchmark models, the ST-DTNN performs higher prediction accuracy during either peak or off-peak periods within an acceptable training time and has more stable prediction performance on the spatial scale. The proposed approach can be extended to develop network-wide traffic state monitoring, optimize routing in navigation services, and support congestion mitigation.

TrafficPredict: Trajectory Prediction for Heterogeneous Traffic-Agents

To safely and efficiently navigate in complex urban traffic, autonomous vehicles must make responsible predictions in relation to surrounding traffic-agents (vehicles, bicycles, pedestrians, etc.). A challenging and critical task is to explore the movement patterns of different traffic-agents and predict their future trajectories accurately to help the autonomous vehicle make reasonable navigation decision. To solve this problem, we propose a long short-term memory-based (LSTM-based) realtime traffic prediction algorithm, TrafficPredict. Our approach uses an instance layer to learn instances’ movements and interactions and has a category layer to learn the similarities of instances belonging to the same type to refine the prediction. In order to evaluate its performance, we collected trajectory datasets in a large city consisting of varying conditions and traffic densities. The dataset includes many challenging scenarios where vehicles, bicycles, and pedestrians move among one another. We evaluate the performance of TrafficPredict on our new dataset and highlight its higher accuracy for trajectory prediction by comparing with prior prediction methods.

Adaptive real time traffic prediction using deep neural networks

<span lang="EN-US">The ever-increasing sale of vehicles and the steady increase in population density in metropolitan cities have raised many growing concerns, most importantly commute time, air and noise pollution levels. Traffic congestion can be alleviated by opting using adaptive traffic light systems, instead of fixed-time traffic signals. In this paper, a system is proposed which can detect, classify and count vehicles passing through any traffic junction using a single camera (as opposed to multi-sensor approaches). The detection and classification are done using SSD Neural Network object detection algorithm. The count of each class (2-wheelers, cars, trucks, buses etc.) is used to predict the signal green-time for the next cycle. The model self-adjusts every cycle by utilizing weighted moving averages. This system works well because the change in the density of traffic on any given road is gradual, spanning multiple traffic stops throughout the day.</span>

Smarter Traffic Prediction Using Big Data, In-Memory Computing, Deep Learning and GPUs.

Road transportation is the backbone of modern economies, albeit it annually costs million deaths and trillions of dollars to the global economy, and damages public health and the environment. Deep learning is among the leading-edge methods used for transportation-related predictions, however, the existing works are in their infancy, and fall short in multiple respects, including the use of datasets with limited sizes and scopes, and insufficient depth of the deep learning studies. This paper provides a novel and comprehensive approach toward large-scale, faster, and real-time traffic prediction by bringing four complementary cutting-edge technologies together: big data, deep learning, in-memory computing, and Graphics Processing Units (GPUs). We trained deep networks using over 11 years of data provided by the California Department of Transportation (Caltrans), the largest dataset that has been used in deep learning studies. Several combinations of the input attributes of the data along with various network configurations of the deep learning models were investigated for training and prediction purposes. The use of the pre-trained model for real-time prediction was explored. The paper contributes novel deep learning models, algorithms, implementation, analytics methodology, and software tool for smart cities, big data, high performance computing, and their convergence.

Real-Time Taxi-Passenger Prediction With L-CNN

The GPS trajectories are rich with potential information that could be used to explore the regulation of traffic to serve the public. While that past approaches for short-term traffic prediction have existed for some time, emerging smart transportation technologies require the traffic prediction capability to be both fast and scalable to full urban networks. In this paper, we propose a novel neural network, named L-CNN based on CNN and LSTM, and develop an effective real-time prediction model to forecast the most likely potential passenger for taxi drivers. It is noteworthy that our model can be easily extended to other real-time traffic prediction problems, such as road traffic and flow prediction. Finally, we test our method based on GPS trajectories generated by Cheng Du taxi. The method presented provides passenger prediction over 15-min intervals for up to 1 h in advance and the results prove the efficiency of our predicting system.

A Two-Level Rolling Optimization Model for Real-time Adaptive Signal Control

Recently, dynamic traffic flow prediction models have increasingly been developed in a connected vehicle environment, which will be conducive to the development of more advanced traffic signal control systems. This paper proposes a rolling optimization model for real-time adaptive signal control based on a dynamic traffic flow model. The proposed method consists of two levels, i.e., barrier group and phase. The upper layer optimizes the length of the barrier group based on dynamic programming. The lower level optimizes the signal phase lengths with the objective of minimizing vehicle delay. Then, to capture the dynamic traffic flow, a rolling strategy was developed based on a real-time traffic flow prediction model. Finally, the proposed method was compared to the Controlled Optimization of Phases (COP) algorithm in a simulation experiment. The results showed that the average vehicle delay was significantly reduced, by as much as 17.95%, using the proposed method.

Efficient Prediction of Network Traffic for Real-Time Applications

Accurate real-time traffic prediction is required in many networking applications like dynamic resource allocation and power management. This paper explores a number of predictors and searches for a predictor which has high accuracy and low computation complexity and power consumption. Many predictors from three different classes, including classic time series, artificial neural networks, and wavelet transform-based predictors, are compared. These predictors are evaluated using real network traces. Comparison of accuracy and cost, both in terms of computation complexity and power consumption, is presented. It is observed that a double exponential smoothing predictor provides a reasonable tradeoff between performance and cost overhead.

Real-Time Traffic Prediction and Probing Strategy for Lagrangian Traffic Data

The objective of this paper is to present a new analytical tool that predicts highway congestion in real time by utilizing a macroscopic traffic flow model, and to investigate a data collection strategy that is adaptable to the quality of traffic information. A stochastic Lagrangian traffic flow model is proposed to capture the transition into traffic jam and randomness in the traffic flow. To calibrate the model, vehicles in a traffic flow are divided into cells, and only the first and last vehicles in each cell are probed. Model parameters and traffic information are updated in real time by the unscented Kalman filter, and an advance warning is provided for stop-and-go traffic jam. Adaptive data collection is done by adjusting the probing cell size based on the variance of the prediction from the stochastic model. By validating the model with empirical highway traffic data, the proposed stochastic model shows a 20% improvement in predicting the one-step-ahead traffic state, comparing it to the result from the deterministic model. The 3-sec prediction of traffic status, which may be applied to compensate for the latency of data processing in real-time applications, can be obtained with a 15% error. The model parameter can be used to warn the drivers 6.76 sec before entering a traffic jam. The scenario with low penetration rate and longer sample time interval is also demonstrated with traffic data collected by smartphones. The results from adaptive probing suggest that it can efficiently use less data to provide higher prediction accuracy than using non-adaptive probing.

Predictive Control for Distributed Smart Street Light Network

With the advent of smart city that embedded with smart technology, namely, smart streetlight, in urban development, the quality of living for citizens has been vastly improved. TALiSMaN is one of the promising smart streetlight schemes to date, however, it possesses certain limitation that led to network congestion and packet dropped during peak road traffic periods. Traffic prediction is vital in network management, especially for real-time decision-making and latency-sensitive application. With that in mind, this paper analyses three real-time short-term traffic prediction models, specifically simple moving average, exponential moving average and weighted moving average to be embedded onto TALiSMaN, that aim to ease network congestion. Additionally, the paper proposes traffic categorisation and packet propagation control mechanism that uses historical road traffic data to manage the network from overload. In this paper, we evaluate the performance of these models with TALiSMaN in simulated environment and compare them with TALiSMaN without traffic prediction model. Overall, weighted moving average showed promising results in reducing the packet dropped while capable of maintaining the usefulness of the streetlight when compared to TALiSMaN scheme, especially during rush hour.

Congestion Prediction With Big Data for Real-Time Highway Traffic

By collecting and analyzing a vast quantity and different categories of information, traffic flow and road congestion can be predicted and avoided in intelligent transportation system. However, how to tackle with these big data is vital but challenging. Most of the existing literatures utilized batch method to process a bunch of road data that cannot achieve real-time traffic prediction. In this paper, we use the spouts and bolts in Apache Storm to implement a real-time traffic prediction model by analyzing enormous streaming data, such as road density, traffic events, and rainfall volume. The proposed SVM-based real-time highway traffic congestion prediction (SRHTCP) model collects the road data from the Taiwan Area National Freeway Bureau, the traffic events reported by road users from the Police Broadcasting Service in Taiwan, and the weather data from the Central Weather Bureau in Taiwan. We use fuzzy theory to evaluate the traffic level of road section in real time with considering road speed, road density, road traffic volume, and the rainfall of road sections. In addition, the SRHTCP model predicts the road speed of next time period by exploring streaming traffic and weather data. Results showed that the proposed SRHTCP model improves 25.6% prediction accuracy than the prediction method based on weighted exponential moving average method under the measurement of mean absolute relative error.

Research on Traffic Speed Prediction by Temporal Clustering Analysis and Convolutional Neural Network With Deformable Kernels (May, 2018)

Real-time traffic speed prediction is one of the most essential parts of the Intelligent Transportation System. In recent years, with the development of artificial intelligence technology, such as in-depth learning, new prediction methods have emerged in endlessly and achieved good results. However, the spatio-temporal information, traffic environment, and their interaction have been hardly depicted in these methods. Therefore, a novel traffic speed prediction method based on temporal clustering analysis and deformable convolution neural network (TCA-DCNN) is proposed in this paper. Temporal clustering analysis (TCA) based on Differential Evolution and hierarchical clustering can adaptively distinguish traffic environment by discriminating the traffic speed variation pattern. By further introducing the deformable convolutional kernels, the characteristics of spatio-temporal traffic speed variation are precisely located in deformable perception fields. After that, a set of DCNN models are trained by using the data processed by TCA, and the output of one of the basic DCNN model is selected as the final traffic speed prediction result. The simulation results based on the measured data in Hangzhou, China, show that the TCA-DCNN algorithm has better performance than other algorithms in traffic speed prediction.

MRF model based real‐time traffic flow prediction with support vector regression

A novel statistical model for real-time traffic flow prediction is proposed. In this study, a 3D Markov random field (MRF) is used to model the temporal dynamics of the traffic flow measured by VDS sensor network. Then, the spatial and temporal relations between roads at a given location are represented by the 3D graph using cliques, then its structure is determined by clique parameters. Here, a support vector regression is adopted for estimating the correlation parameters. The technique is applied to actual traffic flow data from Gyeongbu expressway, South Korea. The experiments demonstrate that the proposed method can predict the traffic flow with an accuracy of 85.6%, which improves 17.3% of the existing state-of-the-art method.

Real-time traffic jams prediction inspired by Biham, Middleton and Levine (BML) model

Urban traffic jams severely affect many cities around the world. Therefore, there is an urgent need for accurate prediction of traffic jams in real time. With many existing approaches, it is difficult to obtain favorable results for dynamic and uncertain traffic flows in urban networks. This paper proposes a real-time traffic jam prediction method based on a two-dimension cellular automaton model inspired by the well-known Biham, Middleton and Levine (BML) model. The proposed model combines the simplicity and high efficiency of the BML model with real urban traffic networks. Our work can be divided into two parts: (1) the proposal of a practical approach for mapping the urban traffic topological structure into a modified BML (M-BML) model, and (2) the development of a novel strategy to deal with different cells of the M-BML model, especially conflict points and fuzzy points, which facilitates the recognition of real traffic jams at intersections. To the best of our knowledge, this is the first time that real urban traffic road networks are directly mapped into a descriptive model for jam prediction. Extensive experiments confirm the accuracy and efficiency of the proposed M-BML model in predicting traffic jams in dynamic and uncertain real urban traffic networks.

Real-time traffic network state estimation and prediction with decision support capabilities: Application to integrated corridor management

This paper presents a real-time traffic network state estimation and prediction system with built-in decision support capabilities for traffic network management. The system provides traffic network managers with the capabilities to estimate the current network conditions, predict congestion dynamics, and generate efficient traffic management schemes for recurrent and non-recurrent congestion situations. The system adopts a closed-loop rolling horizon framework in which network state estimation and prediction modules are integrated with a traffic network manager module to generate efficient proactive traffic management schemes. The traffic network manger adopts a meta-heuristic search mechanism to construct the schemes by integrating a wide variety of control strategies. The system is applied in the context of Integrated Corridor Management (ICM), which is envisioned to provide a system approach for managing congested urban corridors. A simulation-based case study is presented for the US-75 corridor in Dallas, Texas. The results show the ability of the system to improve the overall network performance during hypothetical incident scenarios.

A multi-agent learning approach for online calibration and consistency checking of real-time traffic network management systems

ABSTRACTConsiderable effort has been devoted to developing real-time traffic network management systems for congestion mitigation in large metropolitan areas. These systems usually adopt high-resolution simulation models to provide real-time traffic network state estimation and short-term prediction capabilities. If the simulation results deviate from their corresponding real-world observations, online adjustment to the parameters of the simulation model is recommended to maintain its consistency. For that purpose, several consistency checking and online adjustment modules could be integrated with the simulation model and activated periodically to maintain the model consistency. This paper presents a multi-agent learning methodology for consistency checking and online calibration of real-time traffic network simulation models. The methodology allows multiple online adjustment modules to learn based on their historical performance. Consequently, an adjustment module is activated only if its activation is expected to reduce any detected model inconsistency. The methodology eliminates computation burden associated with the unnecessary activation of these adjustment modules, which might affect the system’s real-time execution requirement. The performance of the methodology is examined using real-world data. The results show that the methodology is promising as an efficient mechanism for maintaining model estimation consistency.

Predictive Unicast and Multicast Scheduling in Onboard Buffered Crossbar Switches

In this letter, a predictive unicast and multicast scheduling (PUMS) scheme is proposed for onboard buffered crossbar switches to cope with time-variant traffic composition. The multicast service ratios required with respect to traffic composition ratio are deduced in theory to guarantee stability and fairness. In the proposed scheme, the multicast service ratios are obtained by a real-time traffic prediction in advance. Accordingly, the bandwidth and crosspoint buffers are adaptively assigned. Simulation results indicate that the proposed PUMS exhibits a promising performance under various traffic scenarios with both fixed and time-variant traffic composition ratios.

Mobile Crowd Sensing for Traffic Prediction in Internet of Vehicles.

The advances in wireless communication techniques, mobile cloud computing, automotive and intelligent terminal technology are driving the evolution of vehicle ad hoc networks into the Internet of Vehicles (IoV) paradigm. This leads to a change in the vehicle routing problem from a calculation based on static data towards real-time traffic prediction. In this paper, we first address the taxonomy of cloud-assisted IoV from the viewpoint of the service relationship between cloud computing and IoV. Then, we review the traditional traffic prediction approached used by both Vehicle to Infrastructure (V2I) and Vehicle to Vehicle (V2V) communications. On this basis, we propose a mobile crowd sensing technology to support the creation of dynamic route choices for drivers wishing to avoid congestion. Experiments were carried out to verify the proposed approaches. Finally, we discuss the outlook of reliable traffic prediction.

Key Technology of Real-Time Road Navigation Method Based on Intelligent Data Research.

The effect of traffic flow prediction plays an important role in routing selection. Traditional traffic flow forecasting methods mainly include linear, nonlinear, neural network, and Time Series Analysis method. However, all of them have some shortcomings. This paper analyzes the existing algorithms on traffic flow prediction and characteristics of city traffic flow and proposes a road traffic flow prediction method based on transfer probability. This method first analyzes the transfer probability of upstream of the target road and then makes the prediction of the traffic flow at the next time by using the traffic flow equation. Newton Interior-Point Method is used to obtain the optimal value of parameters. Finally, it uses the proposed model to predict the traffic flow at the next time. By comparing the existing prediction methods, the proposed model has proven to have good performance. It can fast get the optimal value of parameters faster and has higher prediction accuracy, which can be used to make real-time traffic flow prediction.

Real-time traffic state estimation in urban corridors from heterogeneous data

In recent years, rapid advances in information technology have led to various data collection systems which are enriching the sources of empirical data for use in transport systems. Currently, traffic data are collected through various sensors including loop detectors, probe vehicles, cell-phones, Bluetooth, video cameras, remote sensing and public transport smart cards. It has been argued that combining the complementary information from multiple sources will generally result in better accuracy, increased robustness and reduced ambiguity. Despite the fact that there have been substantial advances in data assimilation techniques to reconstruct and predict the traffic state from multiple data sources, such methods are generally data-driven and do not fully utilize the power of traffic models. Furthermore, the existing methods are still limited to freeway networks and are not yet applicable in the urban context due to the enhanced complexity of the flow behavior. The main traffic phenomena on urban links are generally caused by the boundary conditions at intersections, un-signalized or signalized, at which the switching of the traffic lights and the turning maneuvers of the road users lead to shock-wave phenomena that propagate upstream of the intersections. This paper develops a new model-based methodology to build up a real-time traffic prediction model for arterial corridors using data from multiple sources, particularly from loop detectors and partial observations from Bluetooth and GPS devices.

A novel variable selection method based on frequent pattern tree for real-time traffic accident risk prediction

With the availability of large volumes of real-time traffic flow data along with traffic accident information, there is a renewed interest in the development of models for the real-time prediction of traffic accident risk. One challenge, however, is that the available data are usually complex, noisy, and even misleading. This raises the question of how to select the most important explanatory variables to achieve an acceptable level of accuracy for real-time traffic accident risk prediction. To address this, the present paper proposes a novel Frequent Pattern tree (FP tree) based variable selection method. The method works by first identifying all the frequent patterns in the traffic accident dataset. Next, for each frequent pattern, we introduce a new metric, herein referred to as the Relative Object Purity Ratio (ROPR). The ROPR is then used to calculate the importance score of each explanatory variable which in turn can be used for ranking and selecting the variables that contribute most to explaining the accident patterns. To demonstrate the advantages of the proposed variable selection method, the study develops two traffic accident risk prediction models, based on accident data collected on interstate highway I-64 in Virginia, namely a k-nearest neighbor model and a Bayesian network. Prior to model development, two variable selection methods are utilized: (1) the FP tree based method proposed in this paper; and (2) the random forest method, a widely used variable selection method, which is used as the base case for comparison. The results show that the FP tree based accident risk prediction models perform better than the random forest based models, regardless of the type of prediction models (i.e. k-nearest neighbor or Bayesian network), the settings of their parameters, and the types of datasets used for model training and testing. The best model found is a FP tree based Bayesian network model that can predict 61.11% of accidents while having a false alarm rate of 38.16%. These results compare very favorably with other accident prediction models reported in the literature.