Novel V2X-based traffic congestion prediction system

Because of the population increasing so high, and traffic density remaining the same, traffic prediction has become a great challenge today. Creating a higher degree of communication in automobiles results in the time wastage, fuel wastage, environmental damage, and even death caused by citizens being trapped in the middle of traffic. Only a few researchers work in traffic congestion prediction and control systems, but it may provide less accuracy. So, this paper proposed an efficient IoT-based traffic prediction using OWENN algorithm and traffic signal control system using Intel 80,286 microprocessor for a smart city. The proposed system consists of ‘5’ phases, namely IoT data collection, feature extraction, classification, optimized traffic IoT values, and traffic signal control system. Initially, the IoT traffic data are collected from the dataset. After that, traffic, weather, and direction information are extracted, and these extracted features are given as input to the OWENN classifier, which classifies which place has more traffic. Suppose one direction of the place has more traffic, it optimizes the IoT values by using IBSO, and finally, the traffic is controlled by using Intel 80,286 microprocessor. An efficient OWENN algorithm for traffic prediction and traffic signal control using a Intel 80,286 microprocessor for a smart city. After extracting the features, the classification is performed in this step. Hereabout, the classification is done by using the optimized weight Elman neural network (OWENN) algorithm that classifies which places have more traffic. OWENN attains 98.23% accuracy than existing model also its achieved 96.69% F-score than existing model. The experimental results show that the proposed system outperforms state-of-the-art methods.

A performance modeling and analysis of a novel vehicular traffic flow prediction system using a hybrid machine learning-based model

Traffic prediction system is one of the principal components of an intelligent traffic system (ITS). This system relies on data collected through vehicle-to-vehicle communication, probe vehicle monitoring, speed estimation, and vehicle counting based on vehicle tracking to predict the state of traffic. One of the most demanding tasks in traffic prediction is traffic congestion prediction. Upon predicted traffic congestion, traffic flow control and other intervention can be performed to prevent or at least reduce future congestion, which eases potentially disastrous effects of traffic congestion on the environment, society, and the economy. Traffic congestion prediction is a challenging task in several aspects. First, the prediction needs to be precise. Second, it needs to be made promptly so that any intervention can be meaningful. Third, the system needs the capacity to process a huge amount of data to provide the result for tens of thousands of locations in the map of a city simultaneously. This study proposes a traffic prediction system using Prophet and Spark Streaming. The entire system is built on Apache Spark, which is a Big data processing framework that can be scaled to process a huge amount of data. Spark Streaming is applied to process the streaming data and make real-time forecasting of the traffic flow. The Prophet model, which can capture long-range temporal sequences of data is used to predict traffic flow. The proposed system is shown to achieve good performance based on experimental results with the PEMS-BAY public transport dataset.KeywordsTraffic analysisReal-timeBig dataTime seriesDeep learning

Genetic Network Programming (GNP) is one of the evolutionary optimization algorithms, which uses directed-graph structures to represent its solutions. It has been clarified that GNP works well to find class association rules in traffic prediction systems. In this paper, a novel evolutionary paradigm named GNP with Estimation of Distribution Algorithms (GNP-EDAs) is proposed and used to find important class association rules in traffic prediction systems. In GNP-EDAs, a probabilistic model replaces crossover and mutation to enhance the evolution. The new population of individuals is produced from the probabilistic distribution estimated from the selected elite individuals of the previous generation. The probabilistic information on the connections and transitions of GNP-EDAs is extracted from its population to construct the probabilistic model. In this paper, two methods are described to build the probabilistic model for producing the offspring. In addition, a classification mechanism is introduced to estimate the traffic prediction based on the extracted class association rules. We compared GNPEDAs with the conventional GNP and the simulation results showed that GNP-EDAs can extract the class association rules more effectively, when the number of the candidate class association rules increase. And the classification accuracy of the proposed method shows good results in traffic prediction systems.

The information about Internet traffic should be accurate and timely important for various applications like admission control, congestion control, allocation of bandwidth, and anomaly detection. The prediction of traffic flow is vital for the management and policy of transportation. Mostly, earlier traffic flow prediction techniques utilized simple models for traffic prediction but still these techniques do not meet the desires of various applications of real world. To overcome this, machine learning and fuzzy heterogeneous data sources for Traffic Flow Prediction System (ML-TFPS) is designed and analyzed in this paper. Firstly, the time series model is utilized as a benchmark based on traffic data history for predicting the flow of traffic. Then, heterogeneous data will be integrated for Linear Regression (LR), extreme learning machine (ELM) with machine learning (ML) and fuzzy Traffic Flow Prediction System (MF-TFPS) model. To predict the features of traffic flow, Spark parallelization technology is utilized in described method. MF-TFPS will be intuitively visualized the results of traffic flow prediction. The MF-TFPS will be validated basing on the traffic flow of real data of San Francisco. Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) parameters will be utilized in this study for performance evaluation. From results it is clear that, MF-TFPS with RVM performs well in prediction of traffic flow than the LR, ELM models. The heterogeneous data will be more informative compared to the actual traffic data which is utilized by other researchers, and nonlinear technique utility is demonstrated that can resulting an improvement in the prediction accuracy of traffic flow.

Real-time prediction of traffic congestion enables intelligent transportation systems to improve traffic mobility, reduce delays, and enhance road safety. In this paper, an intelligent traffic congestion prediction system is presented to classify the traffic status across a road network using machine learning. It applies a long short term memory (LSTM) model to estimate the traffic congestion for short term future for LoRa networks. The proposed system, once trained, can efficiently predict traffic congestion using low bandwidth real-time traffic data that can be collected from roadside sensors using a low-power wide area network (LPWAN) technology such as LoRa. We use an online dataset to train and evaluate the proposed model, which shows that the proposed traffic congestion prediction model achieves high performance in terms of accuracy, precision, recall, and success rate. It reduces the error rates and the computing time for fast and accurate future predictions.

Abstract: Most people consider traffic congestion to be a major issue since it increases noise, pollution, and time wastage. Traffic congestion is caused by dynamic traffic flow, which is a serious concern. The current normal traffic light system is not enough to handle the traffic congestion problems since it functions with a fixed-time length strategy. Methods:: Despite the massive amount of traffic surveillance videos and images collected in daily monitoring, deep learning techniques for traffic intelligence management and control have been underutilized. Hence, in this paper, we propose a novel traffic congestion prediction system using a deep learning approach. Initially, the traffic data from the sensors is obtained and pre-processed using normalization. The features are extracted using Multi-Linear Discriminant Analysis (M-LDA). We propose Tri-stage Attention-based Convolutional Neural Network- Recurrent Neural Network (TA- CNN-RNN) for predicting traffic congestion. Results: To evaluate the effectiveness of the proposed model, the Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE) were used as the evaluation metrics. Conclusion: The experimental trial could extend its successful application to the traffic surveillance system and has the potential to enhancement an intelligent transport system in the future.

Machine Learning-based traffic prediction models for Intelligent Transportation Systems

With emerging population and transportation in today’s world, traffic has become a challenging issue to be addressed. Most of the metropolitan cities are facing various traffic-related issues. This poses the need for a smart traffic system, which could tackle the external environment and provide energy efficient transportation system. Intelligent transportation system (ITS) is required to support traffic management system in smart cities. The existing systems concentrate on the traffic prediction to yield better results. The work in this paper proposes a Smart Traffic Prediction and Congestion Avoidance System (s-TPCA) which helps in better identification of the traffic scenario that in turn helps in predicting and avoiding the congestion. The proposed work uses Poisson distribution for prediction of vehicle arrivals from recurring size. The framework comprises traffic identification, prediction, and congestion avoidance phases. The system checks for the fitness function to determine the traffic intensity and further use predictive analytics to determine the traffic level in future. This also integrates fuel consumption model to save time and energy. The proposed s-TPCA system outperforms the conventional systems in terms of delay and proves to conserve energy. The fuel conservation is observed to be 20% higher than the other existing systems.

Hierarchical fuzzy rule-based system optimized with genetic algorithms for short term traffic congestion prediction

Vehicular Clouds introduces a new paradigm that addresses and potentially enhances underutilization of on-board computing resources through aggregation to solve several computational tasks in Intelligent Transportation System. The most challenging issue in Vehicle Cloud is the task allocation among the dynamically changing amount of available resources. For further research towards this issue, a realistic road traffic system models which could generate traffic flow with high accuracy must be designed. In this paper, we conduct a study on short-term traffic flow predictions for our envisioned road traffic prediction system. Five prediction models, including double exponential smoothing (DES), seasonal autoregressive moving average (SARIMA), K-nearest neighbor (KNN), back-propagation neural network (BP-NN) and support vector regression (SVR), are implemented. Then, three different error metrics are used to evaluate the performance of these models. Finally, the results shows that SARIMA and BP neural network are two precise and stationary prediction models and thus are the best candidates to be embedded in an road traffic load prediction system.

Emerging population and transportation in today’s world, traffic has become a challenging issue to be addressed. Most of the metropolitan cities are facing various traffic-related issues. This poses the need for a smart traffic system, which could tackle the external environment and provide energy efficient transportation system. Intelligent transportation system (ITS) is required to support traffic management system in smart cities. The existing systems concentrate on the traffic prediction to yield better results. The work in this paper proposes a Smart Traffic Prediction and Congestion Avoidance System (s-TPCA) which helps in better identification of the traffic scenario that in turn helps in predicting and avoiding the congestion. The proposed work uses Poisson distribution for prediction of vehicle arrivals from recurring size. The framework comprises traffic identification, prediction, and congestion avoidance phases. The system checks for the fitness function to determine the traffic intensity and further use predictive analytics to determine the traffic level in future. This also integrates fuel consumption model to save time and energy. The proposed s-TPCA system outperforms the conventional systems in terms of delay and proves to conserve energy.

Traffic Congestion Prediction using Decision Tree, Logistic Regression and Neural Networks

Aim: The research aims at developing a traffic prediction and signal controlling model based on deep learning technique in order to provide congestion-free transportation in Intelligent Transport System (ITS). Need for the Research: Recent technical advancements in the ITS, industrialization, and urbanization increase traffic congestion, which leads to high fuel consumption and health issues. This signifies the need for a dynamic traffic management system to handle the traffic congestion issues that negatively affect the transportation service. Methods: For promoting congestion-free transportation in the ITS, this research aims to devise a traffic prediction and control system based on deep learning techniques that effectively controls the traffic during peak hours. The proposed mode-search optimization effectively clusters the vehicles based on the necessity. In addition, the mode-search optimization tunes the optimal hyperparameters of the deep Long Short Term Memory classifier, which minimizes the training loss. Further, the traffic signal control system is developed through the mode-search-based deep LSTM classifier for predicting the path of the vehicles by analyzing the attributes, such as velocity, acceleration, jitter, and priority of the vehicles. Result: The experimental results evaluate the efficacy of the traffic prediction model in terms of quadratic mean of acceleration (QMA), jitter, standard deviation of travel time (SDTT), and throughput, for which the values are found to be 37.43, 0.23, 8.75, and 100 respectively. Achievements: The proposed method attains the performance improvement of 5% to 42% when compared with the conventional methods.

IoT devices collect time-series traffic data, which is stochastic and complex in nature. Traffic flow prediction is a thorny task using this kind of data. A smart traffic congestion prediction system is a need of sustainable and economical smart cities. An intelligent traffic congestion prediction model using Seasonal Auto-Regressive Integrated Moving Average (SARIMA) and Bidirectional Long Short-Term Memory (Bi-LSTM) is presented in this study. The novelty of this model is that the proposed model is hybridized using a Back Propagation Neural Network (BPNN). Instead of traditionally presuming the relationship of forecasted results of the SARIMA and Bi-LSTM model as a linear relationship, this model uses BPNN to discover the unknown function to establish a relation between the forecasted values. This model uses SARIMA to handle linear components and Bi-LSTM to handle non-linear components of the Big IoT time-series dataset. The “CityPulse EU FP7 project” is a freely available dataset used in this study. This hybrid univariate model is compared with the single ARIMA, single LSTM, and existing traffic prediction models using MAE, MSE, RMSE, and MAPE as evaluation indicators. This model provides the lowest values of MAE, MSE, RMSE, and MAPE as 0.499, 0.337, 0.58, and 0.03, respectively. The proposed model can help to predict the vehicle count on the road, which in turn, can enhance the quality of life for citizens living in smart cities.