Application of Clustering Algorithm in Intelligent Transportation Data Analysis

Traffic data is a challenging spatio-temporal data, and a multivariate time series data with spatial similarities. Clustering of traffic data is a fundamental tool for various machine learning tasks including anomaly detection, missing data imputation and short term forecasting problems. In this paper, first, we formulate a spatio-temporal clustering problem and define temporal and spatial clusters. Then, we propose an approach for finding temporal and spatial clusters with a deep embedded clustering model. The proposed approach is examined on traffic flow data. In the analysis, we present the properties of clusters and patterns in the dataset. The analysis shows that the temporal and spatial clusters have meaningful relationships with temporal and spatial patterns in traffic data, and the clustering method effectively finds similarities in traffic data.

With the advent of the era of traffic information technology, traffic data appears to be extremely large, and the traditional way of managing highway crashes has become difficult to adapt to the many implications of the complicated data. In this paper, we design a highway traffic safety assessment system based on crash data mining and modeling. By constructing a visualized highway traffic crash data analysis and mining platform, which is used as a basis to gather traffic crash analysis solutions and build a think tank, a safety assessment system is formed that integrates traffic crash information visualization, crash data analysis and mining technology and comprehensive solutions for crash risk prediction. The system provides crash data pre-processing solutions, crash pattern mining solutions, factor analysis solutions and crash risk prediction solutions. Based on the above solution, the basic framework and seven main functional modules of the highway traffic safety assessment system are designed and implemented. The system achieves systematic and automated data processing, pattern mining, factor analysis and risk prediction, and aims to make highway traffic safety management more efficient.

In recent years, tensor has been used in traffic flow data processing. However, the influx of the large number of real-time traffic flow data will cause dimension disaster and computational explosion. In order to deal the problem, this paper presents a Cyber-Physical-Social tensor framework (CPST) and an incremental SVD-based tensor train decomposition algorithm (ITTD), which effectively solves the problem of integrating incremental traffic flow data with original traffic flow data. Firstly, the singular value decomposition (SVD) is used to extract the core tensor of the new traffic flow data, and then the zero-padding operation is performed on the decomposed tensor-train (TT) to make it in the same dimension with the original tensor. After that the new data tensor is fused with the original data tensor. Finally, the fused data tensor train is obtained by orthogonal compression, and the superiority of the algorithm is verified on the UK Highway data set. Our work provides a new data model for traffic flow processing.

Defective traffic flow data may occur because of sensor failure to collect urban traffic data. To solve the issue, a new data mending approach named SARBF neural network fitting is presented. It combines spatial autocorrelation based analysis method and RBF neural network fitting method. In our research, the relevant data-complete intersection need be determined according to the spatial autocorrelation of traffic grid. The historical traffic data of the data-complete intersections could be utilized as training samples of the RBF neural network model. The approach is to mend the defective traffic flow data of the intersections in Hangzhou city. Finally the experiment demonstrated that it can improve the mending precision and computing speed comparing with traditional regression analysis method.

The increasing complexity of urban transportation systems and the growing volume of vehicles have made traffic congestion a persistent challenge in modern cities. Efficient traffic flow prediction is essential for mitigating congestion, improving road safety, optimizing traffic signal control, and enhancing overall transportation efficiency. In recent years, artificial intelligence (AI) has emerged as a transformative tool in the field of traffic management, offering sophisticated algorithms capable of modeling, analyzing, and predicting complex traffic patterns with high accuracy. The application of AI in traffic flow prediction leverages vast amounts of real-time and historical data to generate precise forecasts, supporting data-driven decision-making by urban planners and traffic control authorities. The prediction of traffic flow involves analyzing time-series data that exhibit nonlinear, dynamic, and often stochastic behavior. Traditional statistical models, such as autoregressive integrated moving average (ARIMA), have proven to be limited in handling the high dimensionality and variability inherent in traffic systems. In contrast, AI algorithms possess the capacity to learn and adapt from complex data inputs without the need for explicit programming, making them particularly suitable for traffic-related applications. AI algorithms used in traffic flow prediction can be broadly categorized into machine learning (ML) and deep learning (DL) approaches. Machine learning algorithms such as k-nearest neighbors (KNN), support vector machines (SVM), decision trees, and random forests have demonstrated effectiveness in short-term traffic prediction tasks. These algorithms are capable of identifying hidden patterns in traffic data and adjusting to changes in traffic behavior over time. Ensemble methods, which combine the strengths of multiple learning models, further enhance prediction accuracy and robustness. Deep learning algorithms, a subfield of AI inspired by the human brain’s neural architecture, have shown exceptional performance in capturing spatial-temporal dependencies in traffic data. Recurrent neural networks (RNNs), particularly long short-term memory (LSTM) networks and gated recurrent units (GRUs), are widely used for their ability to process sequential data and retain information over extended time intervals. Convolutional neural networks (CNNs) are employed to extract spatial features from traffic sensor data or road network imagery. Hybrid models that integrate CNNs with RNNs have achieved high levels of predictive precision by simultaneously learning spatial and temporal correlations. In addition to supervised learning methods, unsupervised and reinforcement learning techniques are also applied in traffic flow prediction. Clustering algorithms, such as k-means and DBSCAN, assist in identifying traffic patterns, while reinforcement learning models optimize adaptive traffic signal control systems by learning optimal actions through environmental interaction. This study explores the different types of AI algorithms used in traffic flow prediction, examining their theoretical foundations, structural differences, and practical applications. It aims to evaluate the comparative advantages of various algorithms in addressing the challenges of real-time traffic prediction in increasingly complex transportation networks.

Traffic data analysis is a key step for intelligent transportation systems, and identifying traffic peaks is essential for subsequent traffic pattern analysis. Existing traffic peak detection methods consists of data smoothing and peak picking. We present a max-tree based traffic peak picking method, which constructs the max-tree of the input traffic flow data. Each node in the max tree is a component of an upper level set. We define the prominence of a component as the height difference between a top point and the higher of the left and right foot points of the component. The saliency of a peak is measured by the component prominence. The method generates candidate peaks of positive prominence. The method works directly on noisy traffic data, and the output candidate peaks and their prominences offer the subsequent analysis step the flexibility to choose peaks at any scale.

With the exponential increase in the urban population, urban transportation systems are confronted with numerous challenges. Traffic congestion is common, traffic accidents happen frequently, and traffic environments are deteriorating. To alleviate these issues and improve the efficiency of urban transportation, accurate traffic forecasting is crucial. In this study, we aim to provide a comprehensive overview of the overall architecture of traffic forecasting, covering aspects such as traffic data analysis, traffic data modeling, and traffic forecasting applications. We begin by introducing existing traffic forecasting surveys and preliminaries. Next, we delve into traffic data analysis from traffic data collection, traffic data formats, and traffic data characteristics. Additionally, we summarize traffic data modeling from spatial representation, temporal representation, and spatio-temporal representation. Furthermore, we discuss the application of traffic forecasting, including traffic flow forecasting, traffic speed forecasting, traffic demand forecasting, and other hybrid traffic forecasting. To support future research in this field, we also provide information on open datasets, source resources, challenges, and potential research directions. As far as we know, this paper represents the first comprehensive survey that focuses specifically on the overall architecture of traffic forecasting.

Traffic flow prediction is an important part of intelligent transportation systems (ITS). However, sensor failures or the transmission distortion often occur in the process of data acquisition, which will inevitably cause the loss or abnormality of traffic flow data transmitted to the edge server. In this situation, it is necessary to share traffic flow data among different platforms. However, existing traffic flow prediction methods are facing two challenges in the process of traffic flow data sharing. First, user privacy is often leaked in the process of sharing traffic data on various platforms. Moreover, with the continuous updating of data, the efficiency and scalability of data sharing between different platforms will become lower and lower. In view of the above challenges, in this paper, we propose a novel prediction method for the missing traffic flow data caused by abnormal sensors, named ASMVP_{distr-LSH} based on distributed locality-sensitive hashing (LSH) technique. At last, a case study is presented to illustrate the feasibility and effectiveness of our approach ASMVP_{distr-LSH}.

This study develops cell phone location tracking algorithms from a large cell phone network database to estimate the dynamic origin-destination (O-D) traffic flow and travel demand data as well as commuting traffic. A case study was conducted in the Kansas Metro Corridor to analyze the feasibility of using cell phone data to track cross-region (cities) traffic activities, and to derive the O-D traffic, travel demand by time-of-day and commuting traffic data along the traffic corridor based on a 6 week observation period. The results found that the available cell phone network data detected about 17.6% of the daily traffic data compared to the AADT data along the Kansas Metro Corridor. Approximately 58% of the total traffic was determined to be O-D traffic through the study corridor. This indicates that most of the traffic is from three major regions (the Kansas City metropolitan area, the City of Topeka, KS and the City of Lawrence, KS) and the estimated dynamic travel demand can be used for public transportation system planning and schedule arrangements. Due to the low location resolution using the network-based cell phone network, the use of cell phone network in collecting traffic data would be more feasible for long distance or inter-city trips. A longer observation period is also needed to increase the cell phone sample size and could be useful to obtain stable cell phone traffic, reducing the bias of the data.

Formalizing computational intensity of big traffic data understanding and analysis for parallel computing

Variable message signs (VMSs) provide important traffic information to help drivers travel better on a transportation network. The effectiveness of VMSs largely depends on the numbers and locations of VMSs in a transportation network. Although a few optimization models have been proposed to find candidate roads for locating VMSs, few have been devoted to developing algorithms that can be used in a real transportation network. A large amount of traffic data, such as traffic flow data, is widely available, collected by various means. Based on those traffic data, a sorting algorithm for a VMS location problem is presented in this paper. The algorithm gave a VMS location order rather than a location set. The proposed method divided roads into categories according to multilevel attributes and preferentially selected roads of a higher class with larger flows and more information and the minimal effect of existing VMS in a certain order to locate VMS. The proposed algorithm was analyzed and verified through a practical case on the Beijing, China, urban road network.

Accurate and prompt traffic data are necessary for the successful management of major events. Computer vision techniques, such as convolutional neural network (CNN) applied on video monitoring data, can provide a cost-efficient and timely alternative to traditional data collection and analysis methods. This paper presents a framework designed to take videos as input and output traffic volume counts and intersection turning patterns. This framework comprises a CNN model and an object tracking algorithm to detect and track vehicles in the camera’s pixel view first. Homographic projection then maps vehicle spatial-temporal information (including unique ID, location, and timestamp) onto an orthogonal real-scale map, from which the traffic counts and turns are computed. Several video data are manually labeled and compared with the framework output. The following results show a robust traffic volume count accuracy up to 96.91%. Moreover, this work investigates the performance influencing factors including lighting condition (over a 24-h-period), pixel size, and camera angle. Based on the analysis, it is suggested to place cameras such that detection pixel size is above 2343 and the view angle is below 22°, for more accurate counts. Next, previous and current traffic reports after Texas A&M home football games are compared with the framework output. Results suggest that the proposed framework is able to reproduce traffic volume change trends for different traffic directions. Lastly, this work also contributes a new intersection turning pattern, i.e., counts for each ingress-egress edge pair, with its optimization technique which result in an accuracy between 43% and 72%.

Recent advances in traffic monitoring systems have made real-time traffic velocity data ubiquitously accessible for drivers. This paper develops a traffic data-enabled predictive energy management framework for a power-split plug-in hybrid electric vehicle (PHEV). Compared with conventional model predictive control (MPC), an additional supervisory state of charge (SoC) planning level is constructed based on real-time traffic data. A power balance-based PHEV model is developed for this upper level to rapidly generate battery SoC trajectories that are utilized as final-state constraints in the MPC level. This PHEV energy management framework is evaluated under three different scenarios: 1) without traffic flow information; 2) with static traffic flow information; and 3) with dynamic traffic flow information. Numerical results using real-world traffic data illustrate that the proposed strategy successfully incorporates dynamic traffic flow data into the PHEV energy management algorithm to achieve enhanced fuel economy.

In this paper we present a new approach to gathering traffic data over a continuous-flow of traffic rather than at discrete locations, as is the case with existing technologies. Loop detectors and video cameras, among other devices, currently provide the primary means for gathering data, but the data is gathered at discrete locations. Using cellular technology, the speed and location of each vehicle can be gathered in real-time over a continuous flow, which will allow more novel applications, such as incident identification and probabilities of an incident occurring, to be developed. In addition, as vehicles transmit updated speeds to the system, the fastest path of each commuter from his current location to his destination is maintained. We present and analyze algorithms that can achieve these goals in an efficient manner

With the acceleration of urbanization and the continuous growth of transportation demand, traffic flow prediction has become a key research issue in ITS (Intelligent Transportation System). In this paper, a traffic flow prediction model based on VMD-ARO-BiLSTM is proposed, which can make full use of the spatio-temporal characteristics of traffic data to achieve high-precision traffic flow prediction. First, the traffic flow data are input into VMD (Variational Mode Decomposition) for data reconstruction. Then, the optimal parameter values of BiLSTM (Bi-directional Long Short-Term Memory) are solved using ARO (Artificial Rabbits Optimization) to complete the optimization of BiLSTM. Finally, the reconstructed data are input into BiLSTM to realize the accurate prediction of traffic flow. To verify the model performance, we conducted experiments using traffic flow data from different time periods. The experimental results show that on weekdays, the combined VMD-ARO-BiLSTM prediction model decreases the RMSE (Root Mean Square Error) by 47.2% and the MAE (Mean Absolute Error) by 44.6% compared to BiLSTM; on weekends, compared to BiLSTM, the RMSE decreases by 26.1% and MAE decreases by 23.4%, reflecting the strong robustness and applicability of this model, which has certain application value.