Data-driven Intelligent Transportation Systems Research Articles

AGNP: Network-wide short-term probabilistic traffic speed prediction and imputation

The data-driven Intelligent Transportation System (ITS) provides great support to travel decisions and system management but inevitably encounters the issue of data missing in monitoring systems. Hence, network-wide traffic state prediction and imputation is critical to recognizing the system level state of a transportation network. Abundant research works have adopted various approaches for traffic prediction and imputation. However, previous methods ignore the reliability analysis of the predicted/imputed traffic information. Thus, this study originally proposes an attentive graph neural process (AGNP) method for network-level short-term traffic speed prediction and imputation, simultaneously considering reliability. Firstly, the Gaussian process (GP) is used to model the observed traffic speed state. Such a stochastic process is further learned by the proposed AGNP method, which is utilized for inferring the congestion state on the remaining unobserved road segments. Data from a transportation network in Anhui Province, China, is used to conduct three experiments with increasing missing data ratio for model testing. Based on comparisons against other machine learning models, the results show that the proposed AGNP model can impute traffic networks and predict traffic speed with high-level performance. With the probabilistic confidence provided by the AGNP, reliability analysis is conducted both numerically and visually to show that the predicted distributions are beneficial to guide traffic control strategies and travel plans.

Toward a Novel RESTFUL Big Data-Based Urban Traffic Incident Data Web Service for Connected Vehicles

Abstract Connected vehicles (CVs) are an emerging technology in intelligent transportation systems. Currently, many data-driven intelligent transportation systems (D2ITS) use CV data. Unfortunately, these D2ITS still need serious improvement before they meet higher-level visualization needs. Thus, we aim to develop a new, intelligent data-driven transportation system framework. We focus on visualizing real-time CV data using a big data analytic system in urban areas. In response, we first propose an effective real-time data distribution approach within the Vehicular Ad-hoc NETwork. Second, we develop novel strategies for aggregating, extracting and ingesting data. We provide scalable and fault-tolerant delivery methods without interruption or delay. Finally, we proposed a novel visualization REpresentational State Transfer (REST) web service. We used Simulation of Urban MObility, OMNET++ and Veins to simulate a traffic incident dataset. Then, we tested the Basic Safety Messages in an experimental big data cluster. We used NIFI, Kafka and Cassandra for ingestion, distribution, delivery and storage. The results show accurate performance for packet loss, packet delivery and communication delay. Also, it indicates high throughput and low latency for distributed data delivery systems. Additionally, we obtained the smallest response time for the RESTFUL visualization web service.

Sequence-to-Sequence Recurrent Graph Convolutional Networks for Traffic Estimation and Prediction Using Connected Probe Vehicle Data

Traffic estimation is imperative for conducting fundamental transportation engineering tasks such as transportation planning and traffic safety studies. Additionally, traffic prediction is vital for many data-driven intelligent transportation system applications. Most traffic estimation and prediction methods rely on infrastructure-based sensors to collect traffic parameters. However, infrastructure-based data collection can be costly and time consuming to set up and maintain. Additionally, the spatial distribution of the collected traffic data is limited by the location of the deployed hardware sensors. Probe vehicle data can be used to overcome these limitations. Traffic modeling for estimation and prediction is a complex task due to the stochastic nonlinear spatiotemporal dependencies exhibited by traffic parameters. In this paper, a deep learning-based sequence-to-sequence architecture called Seq2seq GCN-LSTM was proposed to estimate and predict network-wide traffic volume and speed. The proposed method utilizes short-term historical traffic data collected from a low-penetration rate probe vehicle fleet to estimate and predict traffic parameters up to 60-minutes ahead. The method utilizes Graph Convolutional Networks for spatial dependency extraction and Long Short-Term Memory networks to model temporal dependencies. The proposed method generated superior traffic results compared to the baseline models. Additionally, the model demonstrated robustness against perturbations caused by the low rate. Furthermore, the probe vehicle penetration rate was varied to test its effect on the proposed method’s modeling capability. The model was able to maintain traffic volume and speed estimation and prediction performance within a reasonable margin of error using penetration rates as low as 1.5% and 0.5%, respectively.

Traffic Mosaic: Traffic Data Imputation Method Inspired by Image Super Resolution

Missing data is a major problem in data-driven intelligent transportation systems. This study presents an imputation method, called TrafficMosaic, for effective imputation of lost traffic flow data. If we consider traffic data as a image, then the missing data in it can be regarded as a mosaic. The design of TrafficMosaic is inspired by super resolution techniques. First, a deep neural network (DNN) is employed to recover the lost data by a suite of techniques inspired from image super resolution. This DNN method combines the temporal relevance of historical flow with spatial correlation of current fragmented flow data, and uses the imputation result as input for recursive calculation. Second, a data-driven road network matrixing algorithm is proposed to mine location relations from trajectory data and reconstruct a road network flow into matrixes of road network flow pictures. Convolutional local calculation in a convolutional neural network is introduced to extract local spatial features of the road network, thereby reducing the computational complexity and improving the generalization ability. The final step is to extract the local temporal features of the flow data. Traffic flows at adjacent moments have strong temporal characteristics, and we splice several historical road network traffic pictures in chronological order so that our model can capture temporal information from the sequence. The experimental evaluation used a real automatic number plate recognition dataset for experiments, and the results show the effectiveness of TrafficMosaic for lost data imputation.

Prospects and challenges of Metaverse application in data‐driven intelligent transportation systems

The Metaverse is a concept used to refer to a virtual world that exists in parallel to the physical world. It has grown from a conceptual level to having real applications in virtual reality games. The applicability of the Metaverse in numerous sectors like marketing, education, social, and even advertising exists. However, there exists little or no work on Metaverse applicability to the transportation industry. Data-driven intelligent transportation systems (DDITS) aim to provide more intelligent systems based on exploiting data. This paper reviews the concepts and features of the Metaverse. Also, the review goes over three dominant DDITS challenges: vehicle fault detection and repair, testing new technologies, and anti-theft systems. In addition, it highlights prospective Metaverse solutions that apply to the DDITS. Buttressing the utility of Metaverse in DDITS, this paper presents two major case studies: the invisible to visible (I2V) and the Metaverse on Wheels (MoW) technologies. Finally, the influence, limitations, and open issues of Metaverse applications to DDITS are discussed.

A Method for Granular Traffic Data Imputation Based on PARATUCK2

Imputing missing data is a critical task in data-driven intelligent transportation systems. During recent decades there has been a considerable investment in developing various types of sensors and smart systems, including stationary devices (e.g., loop detectors) and floating vehicles equipped with global positioning system (GPS) trackers to collect large-scale traffic data. However, collected data may not include observations from all road segments in a traffic network for different reasons, including sensor failure, transmission error, and because GPS-equipped vehicles may not always travel through all road segments. The first step toward developing real-time traffic monitoring and disruption prediction models is to estimate missing values through a systematic data imputation process. Many of the existing data imputation methods are based on matrix completion techniques that utilize the inherent spatiotemporal characteristics of traffic data. However, these methods may not fully capture the clustered structure of the data. This paper addresses this issue by developing a novel data imputation method using PARATUCK2 decomposition. The proposed method captures both spatial and temporal information of traffic data and constructs a low-dimensional and clustered representation of traffic patterns. The identified spatiotemporal clusters are used to recover network traffic profiles and estimate missing values. The proposed method is implemented using traffic data in the road network of Manhattan in New York City. The performance of the proposed method is evaluated in comparison with two state-of-the-art benchmark methods. The outcomes indicate that the proposed method outperforms the existing state-of-the-art imputation methods in complex and large-scale traffic networks.

Artificial Intelligence-Based Surveillance System for Railway Crossing Traffic

The application of Artificial Intelligence (AI) based techniques has strong potential to improve safety and efficiency in data-driven Intelligent Transportation Systems (ITS) as well as in the emerging Internet of Vehicles (IoV) services. This paper deals with the practical implementation of deep learning methods for increasing safety and security in a specific ITS scenario: railway crossings. This research work presents our proposed system called Artificial Intelligence-based Surveillance System for Railway Crossing Traffic (AISS4RCT) that is based on a combination of detection and classification methods focusing on various image processing inputs: vehicle presence, pedestrian presence, vehicle trajectory tracking, railway barriers at railway crossings, railway warnings, and light signaling systems. The designed system uses cameras that are suitably positioned to capture an entire crossing area at a given railway crossing. By employing GPU accelerated image processing techniques and deep neural networks, the system autonomously detects risky and dangerous situations at railway crossing in real-time. In addition, camera modules send data to a central server for further processing as well as notification to interested parties (police, emergency services, railway operators). Furthermore, the system architecture employs privacy-by-design and security-by-design best practices in order to secure all communication interfaces, protect personal data, and to increase personal privacy, i.e., pedestrians, drivers. Finally, we present field-based results of detection methods, and using the YOLO tiny model method we achieve average recall 89%. The results indicate that our system is efficient for evaluating the occurrence of objects and situations, and it's practicality for use in railway crossings.

IEEE Access Special Section Editorial: Big Data Technology and Applications in Intelligent Transportation

During the last few years, information technology and transportation industries, along with automotive manufacturers and academia, are focusing on leveraging intelligent transportation systems (ITS) to improve services related to driver experience, connected cars, Internet data plans for vehicles, traffic infrastructure, urban transportation systems, traffic collaborative management, road traffic accidents analysis, road traffic flow prediction, public transportation service plan, personal travel route plans, and the development of an effective ecosystem for vehicles, drivers, traffic controllers, city planners, and transportation applications. Moreover, the emerging technologies of the Internet of Things (IoT) and cloud computing have provided unprecedented opportunities for the development and realization of innovative intelligent transportation systems where sensors and mobile devices can gather information and cloud computing, allowing knowledge discovery, information sharing, and supported decision making. However, the development of such data-driven ITS requires the integration, processing, and analysis of plentiful information obtained from millions of vehicles, traffic infrastructures, smartphones, and other collaborative systems like weather stations and road safety and early warning systems. The huge amount of data generated by ITS devices is only of value if utilized in data analytics for decision-making such as accident prevention and detection, controlling road risks, reducing traffic carbon emissions, and other applications which bring big data analytics into the picture.

A Hierarchical Architecture for the Future Internet of Vehicles

Recent advances in wireless communication, sensing, computation and control technologies have paved the way for the development of a new era of Internet of Vehicles (IoV). Demanded by the requirements of information-centric and data-driven intelligent transportation systems (ITS), it is of great significance to explore new paradigms of IoV in supporting large-scale, real-time, and reliable information services. In this article, we propose a hierarchical system architecture, which aims at synthesizing the paradigms of software defined networking and fog computing in IoV and best exploiting their synergistic effects on information services. Specifically, a four-layer architecture is designed, comprising the application layer, the control layer, the virtualization layer, and the data layer, with objectives of enabling logically centralized control via the separation of the control plane and the data plane; facilitating adaptive resource allocation and QoS oriented services based on network functions virtualization and network slicing, and enhancing system scalability, responsiveness, and reliability by exploiting the networking, computation, communication, and storage capacities of fog-based services. On this basis, we further analyze newly arising challenges and discuss future research directions by presenting a cross-layer protocol stack. Finally, for the proof of concept, we implement the system prototype and give two case studies in real-world IoV environments. The results of field tests not only demonstrate the great potential of the new architecture, but also give insight into the development of future ITS.

Missing traffic data imputation and pattern discovery with a Bayesian augmented tensor factorization model

Spatiotemporal traffic data, which represent multidimensional time series on considering different spatial locations, are ubiquitous in real-world transportation systems. However, the inevitable missing data problem makes data-driven intelligent transportation systems suffer from an incorrect response. Therefore, imputing missing values is of great importance but challenging as it is not easy to capture spatiotemporal traffic patterns, including explicit and latent features. In this study, we propose an augmented tensor factorization model by incorporating generic forms of domain knowledge from transportation systems. Specifically, we present a fully Bayesian framework for automatically learning parameters of this model using variational Bayes (VB). Relying on the publicly available urban traffic speed data set collected in Guangzhou, China, experiments on two types of missing data scenarios (i.e., random and non-random) demonstrate that the proposed Bayesian augmented tensor factorization (BATF) model achieves best imputation accuracies and outperforms the state-of-the-art baselines (e.g., Bayesian tensor factorization models). Besides, we discover interpretable patterns from the experimentally learned global parameter, biases, and latent factors that indeed conform to the dynamic of traffic states.

Multi-Agent-Based Data-Driven Distributed Adaptive Cooperative Control in Urban Traffic Signal Timing

Data-driven intelligent transportation systems (D2ITSs) have drawn significant attention lately. This work investigates a novel multi-agent-based data-driven distributed adaptive cooperative control (MA-DD-DACC) method for multi-direction queuing strength balance with changeable cycle in urban traffic signal timing. Compared with the conventional signal control strategies, the proposed MA-DD-DACC method combined with an online parameter learning law can be applied for traffic signal control in a distributed manner by merely utilizing the collected I/O traffic queueing length data and network topology of multi-direction signal controllers at a single intersection. A Lyapunov-based stability analysis shows that the proposed approach guarantees uniform ultimate boundedness of the distributed consensus coordinated errors of queuing strength. The numerical and experimental comparison simulations are performed on a VISSIM-VB-MATLAB joint simulation platform to verify the effectiveness of the proposed approach.

A MapReduce-Based Parallel Frequent Pattern Growth Algorithm for Spatiotemporal Association Analysis of Mobile Trajectory Big Data

Frequent pattern mining is an effective approach for spatiotemporal association analysis of mobile trajectory big data in data-driven intelligent transportation systems. While existing parallel algorithms have been successfully applied to frequent pattern mining of large-scale trajectory data, two major challenges are how to overcome the inherent defects of Hadoop to cope with taxi trajectory big data including massive small files and how to discover the implicitly spatiotemporal frequent patterns with MapReduce. To conquer these challenges, this paper presents a MapReduce-based Parallel Frequent Pattern growth (MR-PFP) algorithm to analyze the spatiotemporal characteristics of taxi operating using large-scale taxi trajectories with massive small file processing strategies on a Hadoop platform. More specifically, we first implement three methods, that is, Hadoop Archives (HAR), CombineFileInputFormat (CFIF), and Sequence Files (SF), to overcome the existing defects of Hadoop and then propose two strategies based on their performance evaluations. Next, we incorporate SF into Frequent Pattern growth (FP-growth) algorithm and then implement the optimized FP-growth algorithm on a MapReduce framework. Finally, we analyze the characteristics of taxi operating in both spatial and temporal dimensions by MR-PFP in parallel. The results demonstrate that MR-PFP is superior to existing Parallel FP-growth (PFP) algorithm in efficiency and scalability.

Spatial-temporal traffic speed patterns discovery and incomplete data recovery via SVD-combined tensor decomposition

Missing data is an inevitable and ubiquitous problem in data-driven intelligent transportation systems. While there are several studies on the missing traffic data recovery in the last decade, it is still an open issue of making full use of spatial-temporal traffic patterns to improve recovery performance. In this paper, due to the multi-dimensional nature of traffic speed data, we treat missing data recovery as the problem of tensor completion, a three-procedure framework based on Tucker decomposition is proposed to accomplish the recovery task by discovering spatial-temporal patterns and underlying structure from incomplete data. Specifically, in the missing data initialization, intrinsic multi-mode biases based traffic pattern is extracted to perform a robust recovery. Thereby, the truncated singular value decomposition (SVD) is introduced to capture main latent features along each dimension. Finally, applying these latent features, the missing data is eventually estimated by the SVD-combined tensor decomposition (STD). Empirically, relying on the large-scale traffic speed data collected from 214 road segments within two months at 10-min interval, our experiment covers two missing scenarios – element-like random missing and fiber-like random missing. The impacts of different initialization strategies for tensor decomposition are evaluated. From numerical analysis, a sensitivity-driven rank selection can not only choose an appropriate core tensor size but also determine how much features we actually need. By comparison with two baseline tensor decomposition models, our method is shown to successfully recover missing data with the highest accuracy as the missing rate ranges from 20% to 80% under two missing scenarios. Moreover, the results have also indicated that an optimal initialization for tensor decomposition could suggest a better performance.

Fusing Incomplete Multisensor Heterogeneous Data to Estimate Urban Traffic

Today, data-driven intelligent transportation systems must address data quality challenges, such as the missing data problem. For example, is it possible to improve the performance of traffic state estimation using incomplete data? In this article, an incomplete traffic data fusing method is proposed to estimate traffic state accurately. It improves missing data estimation by extracting data correlations and applying incomplete data fusion, implementing the two approaches in parallel. The main research focus is on extracting the inherent spatio-temporal correlations of traffic states data from road segments based on a multiple linear regression (MLR) model. Computational experiments for accuracy and efficiency demonstrate that this method can use data correlations to implement accurate and real-time traffic state estimation. This article is part of a special issue on quality modeling.

An Unlicensed Taxi Identification Model Based on Big Data Analysis

Social networks and mobile networks are exposing human beings to a big data era. With the support of big data analytics, conventional intelligent transportation systems (ITS) are gradually changing into data-driven ITS (D2 ITS). Along with traffic growth, D2ITS need to solve more real-life problems, including the issue of unlicensed taxis and their identification, which potentially disrupts the taxi business sector and endangers society safety. As a remedy to this issue, a smart model is proposed in this paper to identify unlicensed taxis. The proposed model consists of two submodel components, namely, candidate selection model and candidate refined model. The former is used to screen out a coarse-grained suspected unlicensed taxi candidate list. The list is taken as an input for the candidate refined model, which is based on machine learning to get a fine-grained list of suspected unlicensed taxis. The proposed model is evaluated using real-life data, and the obtained results are encouraging, demonstrating its efficiency and accuracy in identifying unlicensed taxis, helping governments to better regulate the traffic operation and reduce associated costs.

A distributed spatial–temporal weighted model on MapReduce for short-term traffic flow forecasting

Accurate and timely traffic flow prediction is crucial to proactive traffic management and control in data-driven intelligent transportation systems (D2ITS), which has attracted great research interest in the last few years. In this paper, we propose a Spatial–Temporal Weighted K-Nearest Neighbor model, named STW-KNN, in a general MapReduce framework of distributed modeling on a Hadoop platform, to enhance the accuracy and efficiency of short-term traffic flow forecasting. More specifically, STW-KNN considers the spatial–temporal correlation and weight of traffic flow with trend adjustment features, to optimize the search mechanisms containing state vector, proximity measure, prediction function, and K selection. Furthermore, STW-KNN is implemented on a widely adopted Hadoop distributed computing platform with the MapReduce parallel processing paradigm, for parallel prediction of traffic flow in real time. Finally, with extensive experiments on real-world big taxi trajectory data, STW-KNN is compared with the state-of-the-art prediction models including conventional K-Nearest Neighbor (KNN), Artificial Neural Networks (ANNs), Naïve Bayes (NB), Random Forest (RF), and C4.5. The results demonstrate that the proposed model is superior to existing models on accuracy by decreasing the mean absolute percentage error (MAPE) value more than 11.59% only in time domain and even achieves 89.71% accuracy improvement with the MAPEs of between 3.34% and 6.00% in both space and time domains, and also significantly improves the efficiency and scalability of short-term traffic flow forecasting over existing approaches.

A Survey of Traffic Data Visualization

Data-driven intelligent transportation systems utilize data resources generated within intelligent systems to improve the performance of transportation systems and provide convenient and reliable services. Traffic data refer to datasets generated and collected on moving vehicles and objects. Data visualization is an efficient means to represent distributions and structures of datasets and reveal hidden patterns in the data. This paper introduces the basic concept and pipeline of traffic data visualization, provides an overview of related data processing techniques, and summarizes existing methods for depicting the temporal, spatial, numerical, and categorical properties of traffic data.

A Cyber-ITS Framework for Massive Traffic Data Analysis Using Cyber Infrastructure

Traffic data is commonly collected from widely deployed sensors in urban areas. This brings up a new research topic, data-driven intelligent transportation systems (ITSs), which means to integrate heterogeneous traffic data from different kinds of sensors and apply it for ITS applications. This research, taking into consideration the significant increase in the amount of traffic data and the complexity of data analysis, focuses mainly on the challenge of solving data-intensive and computation-intensive problems. As a solution to the problems, this paper proposes a Cyber-ITS framework to perform data analysis on Cyber Infrastructure (CI), by nature parallel-computing hardware and software systems, in the context of ITS. The techniques of the framework include data representation, domain decomposition, resource allocation, and parallel processing. All these techniques are based on data-driven and application-oriented models and are organized as a component-and-workflow-based model in order to achieve technical interoperability and data reusability. A case study of the Cyber-ITS framework is presented later based on a traffic state estimation application that uses the fusion of massive Sydney Coordinated Adaptive Traffic System (SCATS) data and GPS data. The results prove that the Cyber-ITS-based implementation can achieve a high accuracy rate of traffic state estimation and provide a significant computational speedup for the data fusion by parallel computing.

Data-Driven Intelligent Transportation Systems: A Survey

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.