Learning to Predict Bus Arrival Time From Heterogeneous Measurements via Recurrent Neural Network

Pang, Junbiao; Huang, Jing; Du, Yong; Yu, Haitao; Huang, Qingming; Yin, Baocai

doi:10.1109/tits.2018.2873747

Cited by 60 publications

(49 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, the simple linear relationship could not denote the complexity of real driving environment. Considering the complexity of driving environment, literature [17] proposed heterogeneous impact factor, which considers bus stop location, bus arrival time and departure time, line number of bus stop. It is true that all these factors could be used to express bus real driving state of bus stop, the authors did not consider the impact of real time traffic flow.…”

Section: B Bat Predictionmentioning

confidence: 99%

A Bus Arrival Time Prediction Method Based on Position Calibration and LSTM

et al. 2020

View full text Add to dashboard Cite

Bus arrival time prediction not only provides convenience for passengers, but also helps to improve the efficiency of intelligent transportation system. Unfortunately, the low precision of bus-mounted GPS system, lack of real-time traffic information and poor performance of prediction model lead to low estimation accuracy-greatly influence bus service performance. Hence, in this paper, a GPS calibration method is put forward, while projection rules of specific road shapes are discussed. Moreover, two traffic factors, travel factor and dwelling factor, are defined to express real-time traffic state. Then, considering both historic data and real-time traffic condition, a hybrid dynamic BAT prediction factor, which achieves accuracy enhancement by taking into account traffic flow evaluation results and GPS position calibration, is defined. A LSTM training model is construct to realize BAT prediction. Experiment results demonstrate that our technique can provide a higher level of accuracy compared to methods based on traditional time-ofarrival techniques, especially in the accuracy of multi-stops BAT prediction. INDEX TERMS Bus arrival time prediction, LSTM model, GPS data calibration.

show abstract

Section: B Bat Predictionmentioning

confidence: 99%

A Bus Arrival Time Prediction Method Based on Position Calibration and LSTM

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Incorporating the current location of the bus with the current time and historical data, the model predicts the arrival time of the bus to the next station with an error rate of 12%. It is worth noting that AVL is essential in Kalman Filtering Model in order to estimate the dynamic term in the model [20].…”

Section: Kalman Filtering Modelmentioning

confidence: 99%

Predict Arrival Time by Using Machine Learning Algorithm to Promote Utilization of Urban Smart Bus

Noor¹,

N²,

Kolandaisamy³

et al. 2020

Preprint

View full text Add to dashboard Cite

The impact of the accurate estimated time of arrival (ETA) is often overlooked by bus operators. By providing accurate ETA to riders, it gives them the impression of bus services is efficient and reliable and this promotes higher ridership in the long run. This research project aims to predict bus arrival time by using the Support Vector Regression (SVR) model which is based on the same theory as the Support Vector Machine (SVM). Urban City Bus data covering part of the Petaling Jaya area (route name PJ03) is used in this research work. Features related to traffic such as travel duration, a distance of the road, weather and operation at peak or non-peak hour have been used as input in the training of the SVR model. By using kernel trick and specifying optimum parameters, all the features in higher dimensions are efficiently calculated and the SVR model achieves convergence. The model is evaluated with the test set of data split from the original dataset. The experimental result indicates the SVR model displays good prediction ability with its low average error on the prediction result. However, weather data has not been influential to the prediction model as the results of the model trained with and without weather data show a negligible difference.

show abstract

“…The main difference between LSTM RNN [17] and DA-RNN is that the former does not have the attention mechanism. LSTM RNN is implemented using the PyTorch framework.…”

Section: Lstm Rnnmentioning

confidence: 99%

“…Artificial neural networks have been widely used in various research fields in recent years [13][14][15]. Among artificial neural networks, Multilayer Perceptron (MLP) [16] and Recurrent Neural Network (RNN) [17] have been used to predict bus arrival time. The above methods are of great value to the overall planning of the bus route, but have not yet met the more sensitive time requirements of some tasks such as estimation of passenger waiting time and real-time 2 of 11 scheduling of buses.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Learning Dynamic Factors to Improve the Accuracy of Bus Arrival Time Prediction via a Recurrent Neural Network

et al. 2019

View full text Add to dashboard Cite

Accurate prediction of bus arrival times is a challenging problem in the public transportation field. Previous studies have shown that to improve prediction accuracy, more heterogeneous measurements provide better results. So what other factors should be added into the prediction model? Traditional prediction methods mainly use the arrival time and the distance between stations, but do not make full use of dynamic factors such as passenger number, dwell time, bus driving efficiency, etc. We propose a novel approach that takes full advantage of dynamic factors. Our approach is based on a Recurrent Neural Network (RNN). The experimental results indicate that a variety of prediction algorithms (such as Support Vector Machine, Kalman filter, Multilayer Perceptron, and RNN) have significantly improved performance after using dynamic factors. Further, we introduce RNN with an attention mechanism to adaptively select the most relevant input factors. Experiments demonstrate that the prediction accuracy of RNN with an attention mechanism is better than RNN with no attention mechanism when there are heterogeneous input factors. The experimental results show the superior performances of our approach on the data set provided by Jinan Public Transportation Corporation.

show abstract

Learning to Predict Bus Arrival Time From Heterogeneous Measurements via Recurrent Neural Network

Cited by 60 publications

References 29 publications

A Bus Arrival Time Prediction Method Based on Position Calibration and LSTM

A Bus Arrival Time Prediction Method Based on Position Calibration and LSTM

Predict Arrival Time by Using Machine Learning Algorithm to Promote Utilization of Urban Smart Bus

Learning Dynamic Factors to Improve the Accuracy of Bus Arrival Time Prediction via a Recurrent Neural Network

Contact Info

Product

Resources

About