Multi-Section Traffic Flow Prediction Based on MLR-LSTM Neural Network

Shi, Ruizhe; Du, Lanying

doi:10.3390/s22197517

Cited by 18 publications

(12 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [30], GRU based DNN was implemented with RMSE result of 14.21, 39.99, 27.46, 8.84, and 5.46 in the situations: offpeak, peak, complete dataset, losloop, and Shenzhen road, respectively. In [27], the RMSE for five sections were 31.847, 29.035, 19.352, 68.392, and 81.394, for the first to fifth sections, respectively in the United Kingdom traffic flow dataset, where the LSTM based DNN was implemented. Moreover, in [33] the RMSE obtained by the proposed GRU based DNN was 23.35. however, the suggested DNN in this article achieved RMSEs as 5.56, 5.01, 1.82, and 0.69 for the four-junctions.…”

Section: Resultsmentioning

confidence: 99%

“…The results demonstrate that the suggested method provides estimates of traffic flow that are close to being correct on weekdays and weekends, and under normal and peak traffic situations. Using the continuous and complete traffic flow data in the past period of time of the target prediction section and the incomplete traffic flow data in the past period of time of the target prediction section, a TFP technique utilizing a mix of multiple linear regression (MLR) and LSTM is proposed in [27] to jointly predict the traffic flow modifications of the target section in a short amount of time. When historical data on the intended road segment's traffic flows is incomplete, the aforementioned model can be employed to reliably anticipate how those flows will evolve in the future.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Urban Traffic Flow Estimation System Based on Gated Recurrent Unit Deep Learning Methodology for Internet of Vehicles

et al. 2023

View full text Add to dashboard Cite

Congestion in the world's traffic systems is a major issue that has far-reaching repercussions, including wasted time and money due to longer commutes and more frequent stops for gas. Modern scholarly challenges arise alongside chances to greatly enhance traffic prediction made possible by the integration of modern technologies into transportation systems. Various techniques have been utilized for the purpose of traffic flow prediction, including statistical, machine learning, and deep neural networks. In this paper, deep neural network architecture based on long short term memory (LSTM), bi-directional version, and gated recurrent units (GRUs) layers have been structured to build the deep neural network, in order to predict the performance of the traffic flow in four distinct junctions which has a great impact on the Internet of vehicles' applications. The structure comprised of sixteen-layers, five of them are GRU-layers and one bi-directional LSTM-layer. The dataset employed in this work involved four congested junctions. The dataset extended from the first of November 2016 to 30th of June 2017. Cleaning and preprocessing operations were achieved on the dataset before feeding it to the designed deep neural network of this paper. Results show that the suggested method produced a comparable performance with respect to state-of-the art approaches.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Urban Traffic Flow Estimation System Based on Gated Recurrent Unit Deep Learning Methodology for Internet of Vehicles

et al. 2023

View full text Add to dashboard Cite

show abstract

“…For example, Fu et al [ 24 ] combined LSTM and GRU to predict short-term traffic flow, and Liu et al [ 11 ] combined convolution and LSTM to form a Conv-LSTM model, which can extract spatiotemporal information of the traffic flow information. In addition, Shi et al [ 25 ] proposed a Multiple Linear Regression and a Long Short-Term Memory (MLR-LSTM) model, which uses the incomplete traffic flow data in a past period of the target prediction section and the complete data in a past period of each adjacent section to jointly predict the traffic flow changes of the target section in a short time. Wei et al [ 26 ] proposed a model called AutoEncoder Long Short-Term Memory (AE-LSTM), which uses AutoEncoder to capture the internal relationship of the traffic flow by extracting the characteristics of upstream and downstream traffic flow data and employs LSTM to predict the complex linear traffic flow data.…”

Section: Related Workmentioning

confidence: 99%

Incorporating Multivariate Auxiliary Information for Traffic Prediction on Highways

Xiong

Wan

et al. 2023

Sensors

View full text Add to dashboard Cite

Traffic flow prediction is one of the most important tasks of the Intelligent Transportation Systems (ITSs) for traffic management, and it is also a challenging task affected by many complex factors, such as weather and time. Many cities adopt efficient traffic prediction methods to control traffic congestion. However, most of the existing methods of traffic prediction focus on urban road scenarios, neglecting the complexity of multivariate auxiliary information in highways. Moreover, these methods have difficulty explaining the prediction results based only on the historical traffic flow sequence. To tackle these problems, we propose a novel traffic prediction model, namely Multi-variate and Multi-horizon prediction based on Long Short-Term Memory (MMLSTM). MMLSTM can effectively incorporate auxiliary information, such as weather and time, based on a strategy of multi-horizon time spans to improve the prediction performance. Specifically, we first exploit a multi-horizon bidirectional LSTM model for fusing the multivariate auxiliary information in different time spans. Then, we combine an attention mechanism and multi-layer perceptron to conduct the traffic prediction. Furthermore, we can use the information of multivariate (weather and time) to provide interpretability to manage the model. Comprehensive experiments are conducted on Hangst and Metr-la datasets, and MMLSTM achieves better performance than baselines on traffic prediction tasks.

show abstract

“…The data at the current time and the hidden state at the previous time are input into the LSTM network. The values of the network's input, forgetting and output gates are calculated by fully connected layers with sigmoid activation functions [7] . The flow chart of the LSTM network is shown in Figure 1.…”

Section: Lstm Networkmentioning

confidence: 99%

A short-term highway traffic flow forecasting model based on CNN-LSTM with an attention mechanism

Deng

Sang

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Domestic and foreign scholars have researched traffic flow forecasting and established mature systems; however, these systems are mainly based on normal weather. The influence of weather on traffic flow forecasting is rarely considered in existing studies. Therefore, the correlation between meteorology and traffic flow is analyzed, and an attention mechanism is introduced to improve the ability of the model to extract characteristics of time series data. The results show that the model proposed in this paper is better than other models and the traffic flow forecast using meteorological data is more accurate when an attention mechanism is added.

show abstract

Multi-Section Traffic Flow Prediction Based on MLR-LSTM Neural Network

Cited by 18 publications

References 21 publications

Urban Traffic Flow Estimation System Based on Gated Recurrent Unit Deep Learning Methodology for Internet of Vehicles

Urban Traffic Flow Estimation System Based on Gated Recurrent Unit Deep Learning Methodology for Internet of Vehicles

Incorporating Multivariate Auxiliary Information for Traffic Prediction on Highways

A short-term highway traffic flow forecasting model based on CNN-LSTM with an attention mechanism

Contact Info

Product

Resources

About