Traffic Prediction Problem Research Articles

Bi-State Prediction Network Model for Mixed Traffic Congestion Prediction

Forecasting traffic congestion has become an important issue than ever before. Information provided by traffic congestion prediction methods becomes helpful to traffic management agencies to take suitable actions in due time. On the other hand, with prior information, passengers can plan their trip routes accordingly. In this paper, we tackle the problem of mixed traffic congestion prediction and develop a new Bi-State Prediction Network (BiSPNet). Two heads form BiSPNet. Each head information passed through the layers of three-dimensional convolution (Conv3D), two-dimensional convolutional long short-term memory (ConvLSTM2D), and two-dimensional convolution (Conv2D). Then the dense layer merges both head outputs. These layers capacitate the model to capture the characteristics of multiple inputs. The proposed model is validated by the ground truth data collected by the video cameras installed on the roadways in Delhi, India. Experimental results confirm that the prediction accuracy of the BiSPNet model surpasses that of several state-of-the-art benchmark models. In particular, the BiSPNet reduces mean absolute error by 53.23%, 40.95%, and 0.38%, respectively, in congestion score prediction for 5, 10, and 15-min time horizons compared to the mixed deep learning network (MDLNet). The source data and code can be accessed on GitHub at the following link: https://github.com/amanojup/ .

Successive POI Recommendation via Brain-Inspired Spatiotemporal Aware Representation

Existing approaches usually perform spatiotemporal representation in the spatial and temporal dimensions, respectively, which isolates the spatial and temporal natures of the target and leads to sub-optimal embeddings. Neuroscience research has shown that the mammalian brain entorhinal-hippocampal system provides efficient graph representations for general knowledge. Moreover, entorhinal grid cells present concise spatial representations, while hippocampal place cells represent perception conjunctions effectively. Thus, the entorhinal-hippocampal system provides a novel angle for spatiotemporal representation, which inspires us to propose the SpatioTemporal aware Embedding framework (STE) and apply it to POIs (STEP). STEP considers two types of POI-specific representations: sequential representation and spatiotemporal conjunctive representation, learned using sparse unlabeled data based on the proposed graph-building policies. Notably, STEP jointly represents the spatiotemporal natures of POIs using both observations and contextual information from integrated spatiotemporal dimensions by constructing a spatiotemporal context graph. Furthermore, we introduce a successive POI recommendation method using STEP, which achieves state-of-the-art performance on two benchmarks. In addition, we demonstrate the excellent performance of the STE representation approach in other spatiotemporal representation-centered tasks through a case study of the traffic flow prediction problem. Therefore, this work provides a novel solution to spatiotemporal representation and paves a new way for spatiotemporal modeling-related tasks.

Heterogeneous graph traffic prediction considering spatial information around roads

Precise traffic prediction is crucial in the domain of intelligent transportation. However, the task of accurately predicting traffic has struggled to keep pace with escalating application demands. One of the main reasons for this difficulty is the neglect of the dependence of surrounding spatial data on traffic information. In this study, we introduce a novel framework that incorporates the surrounding spatial data from the road network into the analysis of existing sensor graphs. We delve into an innovative method for traffic prediction by employing a heterogeneous graph that integrates the surrounding spatial information from the road network. The method is based on the common observation that traffic conditions are closely associated with surrounding spatial information, which can be analyzed utilizing heterogeneous graphs. Consequently, we propose a new framework, the heterogeneous attentive spatial–temporal network (HASTN). This framework constructs a heterogeneous graph that merges road networks with surrounding geographic features and employs attention mechanisms to learn traffic patterns. Our method achieves promising results on public datasets as well as a dataset we proposed. Additionally, we employ spatial information to analyze the impact of road traffic patterns on attention. This research offers a fresh perspective on addressing traffic prediction problems by integrating spatial information.

METHOD FOR AGENT-ORIENTED TRAFFIC PREDICTION UNDER DATA AND RESOURCE CONSTRAINTS

Context. Problem of traffic prediction in a city is closely connected to the tasks of transportations in a city as well as air pollution detection in a city. Modern prediction models have redundant complexity when used for separate stations, require large number of measuring stations, long measurement period when predictions are made hourly. Therefore, there is a lack of method to overcome these constraints. The object of the study is a city traffic.
 Objective. The objective of the study is to develop a method for traffic prediction, providing models for traffic quantification at measuring stations in the future under data and resource constraints.
 Method. The method for agent-oriented traffic prediction under data and resource constraints was proposed in the paper. This method uses biLSTM models with input features, including traffic data obtained from agent, representing target station, and other agents, representing informative city stations. These agents are selected by ensembles of decision trees using Random Forest method. Input time period length is proposed to set using autocorrelation data.
 Results. Experimental investigation was conducted on traffic data taken in Madrid from 59 measuring stations. Models created by the proposed method had higher prediction accuracy with lower values of MSE, MAE, RMSE and higher informativeness compared to base LSTM models.
 Conclusions. Obtained models as study results have optimal number of input features compared to the known models, do not require complete system of city stations for all roads. It enables to apply these models under city traffic data and resource constraints. The proposed solutions provide high informativeness of obtained models with practically applicable accuracy level.

Optimized deep extreme learning machine for traffic prediction and autonomous vehicle lane change decision-making

Deep learning occupies a prominent position in the field of machine learning, where researchers have introduced autoencoder into extreme learning machine (ELM) to establish deep extreme learning machine (DELM), with the aim of enabling autonomous feature learning from data. This approach has found widespread applications in various domains, including finance and electrochemistry. However, the network accuracy of DELM is sensitive to the initial weighting and bias configurations, which can be negatively impacted by random initialization. To mitigate the adverse effects of such random initialization on network performance, this study employs two swarm intelligence optimization algorithms, particle swarm optimization (PSO) and harris hawks optimization (HHO), to optimize the input weights and biases of DELM. In addition, we apply the variational mode decomposition (VMD) technique to decompose the original data, enabling a deeper exploration of latent information within the data. Subsequently, two integrated models are constructed: VMD-HHO-DELM and VMD-PSO-DELM. We apply these models to address road traffic condition prediction and autonomous lane change decision-making problems, with a specific focus on the transportation domain. Through extensive experimentation on two real-world datasets, we observe that deep learning models based on swarm intelligence optimization algorithms exhibit excellent performance across various scenarios. In the context of road traffic condition prediction, our experiments reveal superior results under different time intervals and traffic volumes compared to traditional models, all while maintaining higher operational efficiency. For the autonomous lane change decision experiment, we establish a driver dissatisfaction model that aids in formulating more reasonable lane change decisions, thus reducing the frequency of abrupt lane changes.

Location and time embedded feature representation for spatiotemporal traffic prediction

As a fundamental spatiotemporal sequence forecasting problem, traffic prediction is pivotal in transportation management and urban computing. Nonetheless, the intricate and dynamic nature of spatiotemporal correlations presents significant obstacles in acquiring precise forecasts. Existing techniques utilize graph convolutional networks in conjunction with temporal modules, such as recurrent neural networks or transformer-based structures, to effectively extract spatiotemporal features. Unfortunately, current approaches struggle with outliers and fail to capture potential global correlations between different timestamps. In this study, we propose an innovative Spatio-Temporal Graph Convolution Network with Embedded location and time features (STEGCN) for traffic prediction problems, which can generate precise and prompt predictions. STEGCN effectively captures the complex interdependencies among location, time, and traffic volume by leveraging the TransD algorithm to embed their representations. For each timestamp, a graph convolution module is exploited to capture the spatial features, merged with the embeddings of location and time that serve as global external information. Then, we leverage a temporal module composed of 1-D convolutions to capture the spatiotemporal patterns. The traffic volume embedding is employed to constrain predictions within a reasonable range. Extensive experiments and rigorous analysis show that our STEGCN model outperforms state-of-the-art baselines, demonstrating exceptional performance and potential for practical application.

Short-term traffic flow prediction based on weather factors analysis and neural network

With the progress of urban transportation infrastructure and the construction of intelligent transportation system, sufficiently accurate traffic flow prediction is of increasing importance for urban planning and management. This paper based on traffic flow data in a section of highway in Los Angeles, USA, we combine external qualitative data such as weather and work calendars, analyze and process the factors significantly influencing traffic flow first, then train and test the traffic flow data by building machine learning Support Vector Regression (SVR) model and the Back Propagation neural network (BP neural network). The test results show that the BP neural network traffic flow prediction model incorporating the external environment is more accurate, and that the external environment has a significant impact on the traffic flow prediction problem. If more significantly influencing factors can be taken into account, or if the current algorithm can be optimized, the prediction accuracy will have further improved, thus effectively enhancing the efficiency and safety of traffic systems.

RL-GCN: Traffic flow prediction based on graph convolution and reinforcement learning for smart cities

The traffic flow problem has become essential in urban planning and management in today’s increasingly urbanized world. Traditional traffic flow prediction models cannot fully consider urban traffic networks’ complex and dynamic characteristics. To this end, this paper proposes a traffic flow prediction method for smart cities (RL-GCN) based on graph convolution, LSTM network and reinforcement learning, aiming to solve the problem of urban traffic flow prediction. Firstly, we use the graph convolutional neural network to process the urban traffic network data features, then use the LSTM network model to learn the temporal information, and then combine the reinforcement learning algorithm to develop the optimal traffic control strategy based on which the future traffic flow is predicted. Our experiments on several datasets show that the model developed in this paper has outstanding performance for urban traffic flow prediction. Compared with the traditional traffic flow prediction methods, the method in this paper has significantly improved prediction accuracy. Our research can provide valuable references and inspiration in urban planning and traffic management.

Few-Shot traffic prediction based on transferring prior knowledge from local network

Short-term traffic prediction has been widely studied in the community of Intelligent Transport Systems for decades. Despite the advances in machine learning-based prediction techniques, a challenging problem that affects the applications of such methods in practice is the prevalence of insufficient data across an entire road network. To address this few-shot traffic prediction problem at a local network scale, we develop a hybrid framework in conjunction with the prior knowledge transferring algorithm and two widely used models, i.e. Long-short Term Memory and Spatial–Temporal Graph Convolutional Neural Network. The proposed modelling framework is trained and tested using five-minute interval traffic flow data collected from London under different few-shot learning scenarios. Results show that transferring local network prior knowledge can improve the accuracy of both one-step prediction and multi-step prediction under inadequate data conditions, regardless of the deep-learning tool used.

Knowledge Expansion and Consolidation for Continual Traffic Prediction With Expanding Graphs

Accurate traffic prediction plays a vital role in intelligent transport managements and applications. However, in the vast majority of existing works, the focus is mainly on modeling spatiotemporal correlations in static traffic networks. Thus, the continuous expansion and evolution of traffic networks are ignored. In this work, we study the problem of traffic prediction with expanding road network structures under the continual learning paradigm. Considering the model prediction performance, efficiency, and data accessibility, a SpatioTemporal Knowledge Expansion and Consolidation (STKEC) framework is proposed. This framework contains an influence-based knowledge expansion strategy to help the spatiotemporal learning model integrate new spatiotemporal traffic patterns and a memory-augmented knowledge consolidation mechanism to preserve the learned spatiotemporal patterns without accessing the data in previous graphs. Extensive experiments are conducted on a large-scale dataset and verify the superior performance of STKEC in continual traffic prediction.

Trafformer: Unify Time and Space in Traffic Prediction

Traffic prediction is an important component of the intelligent transportation system. Existing deep learning methods encode temporal information and spatial information separately or iteratively. However, the spatial and temporal information is highly correlated in a traffic network, so existing methods may not learn the complex spatial-temporal dependencies hidden in the traffic network due to the decomposed model design. To overcome this limitation, we propose a new model named Trafformer, which unifies spatial and temporal information in one transformer-style model. Trafformer enables every node at every timestamp interact with every other node in every other timestamp in just one step in the spatial-temporal correlation matrix. This design enables Trafformer to catch complex spatial-temporal dependencies. Following the same design principle, we use the generative style decoder to predict multiple timestamps in only one forward operation instead of the iterative style decoder in Transformer. Furthermore, to reduce the complexity brought about by the huge spatial-temporal self-attention matrix, we also propose two variants of Trafformer to further improve the training and inference speed without losing much effectivity. Extensive experiments on two traffic datasets demonstrate that Trafformer outperforms existing methods and provides a promising future direction for the spatial-temporal traffic prediction problem.

VDGCNeT: A novel network-wide Virtual Dynamic Graph Convolution Neural network and Transformer-based traffic prediction model

We address the problem of traffic prediction on large-scale road networks. We propose a novel deep learning model, Virtual Dynamic Graph Convolution Neural Network and Transformer with Gate and Attention mechanisms (VDGCNeT), to comprehensively extract complex, dynamic and hidden spatial dependencies of road networks for achieving high prediction accuracy. For this purpose, we advocate the use of a virtual dynamic road graph that captures the dynamic and hidden spatial dependencies of road segments in real road networks instead of purely relying on the physical road connectivity. We further design a novel framework based on Graph Convolution Neural Network (GCN) and Transformer to analyze dynamic and hidden spatial–temporal features. The gate mechanism is utilized for concatenating learned spatial and temporal features from Spatial and Temporal Transformers, respectively, while the Attention-based Similarity is used to update dynamic road graph. Two real-world traffic datasets from large-scale road networks with different properties are used for training and testing our model. We compare our VDGCNeT against nine other well-known models in the literature. Our results demonstrate that the proposed VDGCNeT is capable of achieving highly accurate predictions — on average 96.77% and 91.68% accuracy on PEMS-BAY and METR-LA datasets respectively. Overall, our VDGCNeT performs the best when compared against other existing models.

Using neural networks to explore path planning algorithms for robots

In the introduction section of this paper, the importance of path planning for automatic self-navigation is outlined, along with the structure and basic model of neural networks, as well as numerous practical examples of using neural networks to solve the problem of traffic prediction and autonomous robot navigation. In the method section, the specific formulations and steps of how neural networks can be used to solve the path planning problem are explained in detail. Additionally, a variety of algorithms for solving the path planning problem with neural networks are introduced, and practical experiments are conducted to compare the results of these algorithms in various environments. In the RESULTS section, the characteristics are summarized and explained by comparing the experimental results. In the CONFUSION section, a fundamental outlook on the future of neural networks for solving path-planning problems is provided, as well as the possibility of cooperation between neural networks and other methods for solving path-planning problems.

Lane-Level Heterogeneous Traffic Flow Prediction: A Spatiotemporal Attention-Based Encoder–Decoder Model

Urban road traffic flow prediction is the key basis for the development of intelligent transportation systems. Lane-level heterogeneous traffic flow prediction will become a new and important challenge in the future development of intelligent transportation. In this study, a spatiotemporal attention-based encoder-decoder model is proposed to solve the lane-level heterogeneous traffic flow prediction problem. Specifically, the traffic flows of different locations and different types of vehicles are all regarded as traffic flows running in independent spaces. A spatial attention layer is added to the encoder to capture the spatial characteristics of traffic flows, and a temporal attention layer is added to the decoder to analyze the temporal correlation of each time interval. Finally, we validate the model by using real traffic monitoring data in the Shunyi District, Beijing, China, and prove the superior performance of the model. Meanwhile, the output of attention weights can deepen the further interpretation of the spatiotemporal correlation of heterogeneous traffic flows.

Exploring the Generalizability of Spatio-Temporal Traffic Prediction: Meta-Modeling and an Analytic Framework

The Spatio-Temporal Traffic Prediction (STTP) problem is a classical problem with plenty of prior research efforts that benefit from traditional statistical learning and recent deep learning approaches. While STTP can refer to many real-world problems, most existing studies focus on quite specific applications, such as the prediction of taxi demand, ridesharing order, traffic speed, and so on. This hinders the STTP research as the approaches designed for different applications are hardly comparable, and thus how an application-driven approach can be generalized to other scenarios is unclear. To fill in this gap, this paper makes three efforts: (i) we propose an analytic framework, called STAnalytic, to qualitatively investigate STTP approaches regarding their design considerations on various spatial and temporal factors, aiming to make different application-driven approaches comparable; (ii) we design a spatio-temporal meta-model, called STMeta, which can flexibly integrate generalizable temporal and spatial knowledge identified by STAnalytic, (iii) we build an STTP benchmark platform including ten real-life datasets with five scenarios to quantitatively measure the generalizability of STTP approaches. In particular, we implement STMeta with different deep learning techniques, and STMeta demonstrates better generalizability than state-of-the-art approaches by achieving lower prediction error on average across all the datasets.

A hybrid model for metro passengers flow prediction

In this paper, a novel ensemble learning model named EWT-EnsemLSTM-SSA, which assembles long short-term memory (LSTM), support vector regression (SVR), and sparrow search algorithm (SSA), is a proposed to deal with long term metro passenger flow volume prediction, which is an essential content of traffic flow prediction problems. Firstly, the empirical wavelet transform (EWT) method is introduced to decompose the original dataset into five wavelet time-sequence data for further prediction. Then, a cluster of LSTMs with diverse hidden layers and neuron counts are employed to explore and exploit the implicit information of the EWT-decomposed signals. Next, the output of LSTMs is aggregated into a nonlinear regression method SVR. Lastly, SSA is utilized to optimize the SVR automatically. The proposed EWT-EnsemLSTM-SSA model is applied in three case studies, using the data collected from the passengers’ amount in the Minneapolis, America metro, divided into one hour in one day. Experiment results, which compare the proposed EnsemLSTM-SSA model with five conventional time series forecasting models, show that the proposed model can achieve a better performance than the traditional prediction algorithms.

Uncertainty quantification and consideration in ML-aided traffic-driven service provisioning

The network traffic prediction problem has been extensively studied in the literature through the application of statistical linear models and more recently through the application of machine learning (ML). In fact, ML has proven its capabilities on accurately modeling the non-linear nature of network traffic, outperforming conventional statistical linear models. Without doubt, model accuracy constitutes an important evaluation metric. In a network environment, however, where uncertainty may lead to erroneous service provisioning decisions (e.g., violations of the quality-of-service (QoS) requirements), it does not provide any information of how much the traffic predictions can be trusted. Hence, in this work the focus is on addressing traffic prediction uncertainty by leveraging the capabilities of Monte Carlo (MC) dropout inference. The proposed framework is compared with a margin-based technique, traditionally used to compensate for traffic prediction uncertainty, demonstrating that the MC dropout inference framework results in significant spectrum savings, as opposed to the baseline, myopic, margin-based scheme. Even though a small penalty is observed on the unpredictable traffic, this is successfully handled on-line with a reduced operational overhead compared to the case where prediction uncertainty is completely ignored. Importantly, it is shown that, unlike the margin-based framework, the MC dropout inference framework can be used for the provisioning of services with diverse QoS requirements.

RPConvformer: A novel Transformer-based deep neural networks for traffic flow prediction

Traffic prediction problem is one of the essential tasks of intelligent transportation system (ITS), alleviating traffic congestion effectively and promoting the intelligent development of urban traffic. To accommodate long-range dependencies, Transformer-based methods have been used in traffic prediction tasks due to the parallelizable processing of sequences and explanation of attention matrices compared with recurrent neural units (RNNs). However, the Transformer-based model has two limitations, on the one hand, it ignores the local correlation in the traffic state in its parallel processing of the sequence, on the other hand, the absolute positional embedding is adopted to represent the positional relationship of time nodes is destroyed when it comes to calculate attention score. To address two embarrassing shortcomings, a novel framework called RPConvformer is proposed, where the improved parts are 1D causal convolutional sequence embedding and relative position encoding. In sequence embedding, we develop a sequence embedding layer composed of convolutional units, which consist of origin 1D convolutional and 1D causal convolutional. The size of the receptive field of the convolution can focus on the local region correlation of the sequence. In relative position encoding, we introduce a bias vector to automatically learn the relative position information of time nodes when linearly mapping the feature tensor. We respect the encoding and decoding framework of the Transformer, the encoder is responsible for extracting historical traffic state information, and the decoder autoregressively predicts the future traffic state. The multi-head attention mechanism is adopted by both encoder and decoder aims to focus on rich temporal feature patterns. Moreover, key mask technique is used after computing attention matrix to mask the traffic state at missing moments improving the resilience of the model. Extensive experiments on two real-world traffic flow datasets. The results show that RPConvformer achieves the best performance compared to state-of-the-art time series models. Ablation experiments show that considering the local correlation of time series has a higher gain on prediction performance. Random mask experiments show that the model is robust when the historical data is less than 10% missing. In addition, multi-head attention matrix provides further explanation for the dependence between time nodes. RPConvformer as an improved Transformer-based model can provide new ideas for molding temporal dimension in traffic prediction tasks. Our code has been open-sourced at (https://github.com/YanJieWen/RPConvformer).

Short-Term Traffic Prediction on Swedish Highways: A Deep Learning Approach with Knowledge Representation

Accurate prediction of highway traffic is of vital importance to proactive traffic monitoring, operation and controls. In the data mining of highway traffic, abstracting temporal knowledge is often prioritized than exploring topological relationship. In this study, we propose a deep learning model, called Knowledge-Sequence-to-Sequence (K-Seq2Seq), to solve the short-term highway traffic prediction problem in two stages: representing temporal knowledge and predicting future traffic. Through computational experiment in a road section of a Swedish motorway, we show that our model outperforms the conventional Seq2Seq model significantly, more than 20% when predicting information of longer time step.

A Cyberphysical System for Data-Driven Real-Time Traffic Prediction on the Las Vegas I-15 Freeway

Mobility and transportation services in modern large-scale cities face traffic congestion as one of the main sources of discomfort and economic losses. In this context, taking preventive measures based on traffic predictions looks like an appealing alternative to mitigate congestion. The increasing availability of detectors in the transportation infrastructure has allowed tackling the traffic prediction problem by using a purely data-driven approach, where deep learning models have excelled. Unfortunately, the implementation of these techniques in real time is still under development. This work presents the implementation of a real-time traffic prediction application in the Las Vegas, NV, USA, urban area, built as a cyberphysical system with real-time data streaming from field sensors to a cloud-like environment where deep learning-based traffic predictors are hosted. Implementation results show the feasibility of doing traffic prediction in real time with the current technology and the usefulness of periodic retraining to maintain prediction accuracy.