Bus travel time prediction with real-time traffic information
TL;DR: This work proposes a novel segment-based approach to predict bus travel times using a combination of real-time taxi and bus datasets, that can automatically divide bus routes into dwelling and transit segments and improves the accuracy of bus travel time prediction.
Abstract: An important aspect of Intelligent Public Transportation Systems (IPTS) is providing accurate travel time information. Knowing arrival times of public vehicles in advance can reduce waiting times of passengers and attract more people to take public transport. Existing approaches have two main limitations in the field of bus travel time prediction. First, influenced by increasingly complex real-time traffic factors and sparsity of real-time data, bus travel times can be difficult to predict accurately in modern cities. Second, bus dwelling and transit times are predominantly affected by different factors and hence have different patterns, but little research focuses on how to divide dwelling and transit areas and to build independent models for them. Consequently, we propose a novel segment-based approach to predict bus travel times using a combination of real-time taxi and bus datasets, that can automatically divide bus routes into dwelling and transit segments. Two models are built to predict them separately by incorporating different impact traffic factors. We evaluate our approach using real-world trajectory datasets, collected in Xi’an, China during June 2017. Compared to existing methods, the experimental results reveal that our approach improves the accuracy of bus travel time prediction, especially under abnormal traffic conditions.
Citations
More filters
TL;DR: The end-to-end trainable RSTN redefines the traditional prediction problem as a learning residual function with regard to the travel density in each time interval, and a dynamic request vector (DRV)-based data representation scheme is presented, to improve the performance of forecasting.
Abstract: This paper proposes a deep architecture called residual spatio-temporal network (RSTN) for short-term travel demand forecasting. It comprises fully convolutional neural networks (FCNs) and a hybrid module consisting of an extended Conv-LSTM (CE-LSTM) that can achieve trade-off of convolutional operation and LSTM cells by tuning the hyperparameters of Conv-LSTM, convolutional neural networks (CNNs) and traditional LSTM. These modules are combined via residual connections to capture the spatial, temporal and extraneous dependencies of travel demand. The end-to-end trainable RSTN redefines the traditional prediction problem as a learning residual function with regard to the travel density in each time interval. Further more, a dynamic request vector (DRV)-based data representation scheme is presented, which catches the intrinsic characteristics and variation of the trend, to improve the performance of forecasting. Simulations with two real-word data sets show that the proposed method outperforms the existing forecasting algorithms, reducing the root mean square error (RMSE) by up to 17.87%.
71 citations
TL;DR: The Back Propagation (BP) neural network model can be used to predict the passage time of vehicles queuing at intersections with an error of less than 10%, and the improved CAN bus communication can improve the data transmission rate.
Abstract: Intelligent control of traffic has significant influence on the scheduling efficiency of urban traffic flow. Therefore, in order to improve the efficiency of vehicles at intersections, first, the Back Propagation (BP) neural network is used to propose a vehicle passing model at the intersection, and based on the intelligent traffic control system model, the Earliest Deadline First (EDF) dynamic scheduling algorithm is used to improve the Controller Area Network (CAN) communication network. Finally, the simulation test is used to evaluate the effectiveness of the proposed model and the improved CAN bus communication network. The results show that the neural network model can be used to predict the passage time of vehicles queuing at intersections with an error of less than 10%. The improved CAN bus communication can improve the data transmission rate, and the success rate of data transmission under different load rates is above 95%. In conclusion, the application of artificial intelligence technology in intelligent traffic system can improve the efficiency of vehicle scheduling and the efficiency of communication system. This research is of great significance to improve the communication performance of the transportation system and scheduling efficiency.
66 citations
TL;DR: A long short-term memory (LSTM) and Artificial neural networks (ANN) comprehensive prediction model based on spatial-temporal features vectors is proposed to solve the problems of remote dependence of bus arrival and road incidents and has high accuracy among bus arrival prediction problems.
Abstract: Bus arrival prediction has important implications for public travel, urban dispatch, and mitigation of traffic congestion. The factors affecting urban traffic conditions are complex and changeable. As the predicted distance increases, the difficulty of traffic prediction becomes more difficult. Forecast based on historical data responds quite slowly for changes under the short-term conditions, and vehicle prediction based on real-time speed is not sufficient to predict under long-term conditions. Therefore, an arrival prediction method based on temporal vector and another arrival prediction method based on spatial vector is proposed to solve the problems of remote dependence of bus arrival and road incidents, respectively. In this paper, combining the advantages of the two prediction models, this paper proposes a long short-term memory (LSTM) and Artificial neural networks (ANN) comprehensive prediction model based on spatial-temporal features vectors. The long-distance arrival-to-station prediction is realized from the dimension of time feature, and the short-distance arrival-to-station prediction is realized from the dimension of spatial feature, thereby realizing the bus-to-station prediction. Besides, experiments were conducted and tested based on the entity dataset, and the result shows that the proposed method has high accuracy among bus arrival prediction problems.
42 citations
TL;DR: This paper proposes a novel urban flow prediction framework by generalizing the hidden states of the model with continuous-time dynamics of the latent states using neural ordinary differential equations (ODE), and introduces a discretize-then-optimize approach to improve and balance the prediction accuracy and computational efficiency.
Abstract: With the recent advances in deep learning, data-driven methods have shown compelling performance in various application domains enabling the Smart Cities paradigm. Leveraging spatial–temporal data from multiple sources for (citywide) traffic forecasting is a key to strengthen the smart city management in areas such as urban traffic control, abnormal event detection, etc. Existing approaches of traffic flow prediction mainly rely on the development of various deep neural networks –e.g., Convolutional Neural Networks such as ResNet are used for modeling spatial dependencies among different regions, whereas recurrent neural networks are increasingly implemented for temporal dynamics modeling. Despite their advantages, the existing approaches suffer from limitations of intensive computations, lack of capabilities to properly deal with missing values, and simplistic integration of heterogeneous data. In this paper, we propose a novel urban flow prediction framework by generalizing the hidden states of the model with continuous-time dynamics of the latent states using neural ordinary differential equations (ODE). Specifically, we introduce a discretize-then-optimize approach to improve and balance the prediction accuracy and computational efficiency. It not only guarantees the prediction error but also provides high flexibility for decision-makers. Furthermore, we investigate the factors, both intrinsic and extrinsic, that affect the city traffic volume and use separate neural networks to extract and disentangle the influencing factors, which avoids the brute-force data fusion in previous works. Extensive experiments conducted on the real-world large-scale datasets demonstrate that our method outperforms the state-of-the-art baselines, while requiring significantly less memory cost and fewer model parameters.
27 citations
01 Aug 2020
TL;DR: DeepTRANS is presented, which incorporates traffic forecasting information to the authors' prior Deep Learning-based Bus Estimated Time of Arrival (ETA) model, increasing its accuracy by 21% in estimating bus travel time.
Abstract: In the public transportation domain, accurate estimation of travel times helps to manage rider expectations as well as to provide a powerful tool for transportation agencies to coordinate the public transport vehicles. Although many statistical and machine learning methods have been proposed to estimate travel times, none of the methods consider utilizing predicted traffic information. Forecasting how congestion is going to evolve is critical for accurate travel time estimations. In this paper, we present DeepTRANS, which incorporates traffic forecasting information to our prior Deep Learning-based Bus Estimated Time of Arrival (ETA) model, increasing its accuracy by 21% in estimating bus travel time.
19 citations
Cites background or methods from "Bus travel time prediction with rea..."
...Many statistical and machine learning methods have been proposed to estimate travel times, e.g., Support Vector Machine [5] and Recurrent Neural Network [6, 2]....
[...]
...While most recent methods strive to improve their estimation by incorporating current traffic situation [5], none have utilized the predicted traffic information....
[...]
...We concatenate the traffic flow predictions produced by our DCRNN model with the output vectors of the Geo-Conv layer in DeepTTE....
[...]
..., Support Vector Machine [5] and Recurrent Neural Network [6, 2]....
[...]
...For both cases, we compare the estimated times of arrival from our Bus ETA model with those from the deep learning model, DeepTTE, and statistical model, GBDT. Scenario 1: Current or Future Bus Arrival Time Estimation....
[...]
References
More filters
Proceedings Article•
02 Aug 1996TL;DR: In this paper, a density-based notion of clusters is proposed to discover clusters of arbitrary shape, which can be used for class identification in large spatial databases and is shown to be more efficient than the well-known algorithm CLAR-ANS.
Abstract: Clustering algorithms are attractive for the task of class identification in spatial databases. However, the application to large spatial databases rises the following requirements for clustering algorithms: minimal requirements of domain knowledge to determine the input parameters, discovery of clusters with arbitrary shape and good efficiency on large databases. The well-known clustering algorithms offer no solution to the combination of these requirements. In this paper, we present the new clustering algorithm DBSCAN relying on a density-based notion of clusters which is designed to discover clusters of arbitrary shape. DBSCAN requires only one input parameter and supports the user in determining an appropriate value for it. We performed an experimental evaluation of the effectiveness and efficiency of DBSCAN using synthetic data and real data of the SEQUOIA 2000 benchmark. The results of our experiments demonstrate that (1) DBSCAN is significantly more effective in discovering clusters of arbitrary shape than the well-known algorithm CLAR-ANS, and that (2) DBSCAN outperforms CLARANS by a factor of more than 100 in terms of efficiency.
17,056 citations
Proceedings Article•
01 Jan 1996TL;DR: DBSCAN, a new clustering algorithm relying on a density-based notion of clusters which is designed to discover clusters of arbitrary shape, is presented which requires only one input parameter and supports the user in determining an appropriate value for it.
Abstract: Clustering algorithms are attractive for the task of class identification in spatial databases. However, the application to large spatial databases rises the following requirements for clustering algorithms: minimal requirements of domain knowledge to determine the input parameters, discovery of clusters with arbitrary shape and good efficiency on large databases. The well-known clustering algorithms offer no solution to the combination of these requirements. In this paper, we present the new clustering algorithm DBSCAN relying on a density-based notion of clusters which is designed to discover clusters of arbitrary shape. DBSCAN requires only one input parameter and supports the user in determining an appropriate value for it. We performed an experimental evaluation of the effectiveness and efficiency of DBSCAN using synthetic data and real data of the SEQUOIA 2000 benchmark. The results of our experiments demonstrate that (1) DBSCAN is significantly more effective in discovering clusters of arbitrary shape than the well-known algorithm CLARANS, and that (2) DBSCAN outperforms CLARANS by a factor of more than 100 in terms of efficiency.
14,297 citations
TL;DR: There are limitations, mainly that trip length is not recorded on systems based on validating cards only on entry to a bus, and that certain types of data still require direct survey methods for their collection (such as journey purpose).
386 citations
TL;DR: Two artificial neural networks, trained by link-based and stop-based data, are applied to predict transit arrival times and show that the enhanced ANNs outperform the ones without integration of the adaptive algorithm.
Abstract: Transit operations are interrupted frequently by stochastic variations in traffic and ridership conditions that deteriorate schedule or headway adherence and thus lengthen passenger wait times. Providing passengers with accurate vehicle arrival information through advanced traveler information systems is vital to reducing wait time. Two artificial neural networks (ANNs), trained by link-based and stop-based data, are applied to predict transit arrival times. To improve prediction accuracy, both are integrated with an adaptive algorithm to adapt to the prediction error in real time. The bus arrival times predicted by the ANNs are assessed with the microscopic simulation model CORSIM, which has been calibrated and validated with real-world data collected from route number 39 of the New Jersey Transit Corporation. Results show that the enhanced ANNs outperform the ones without integration of the adaptive algorithm.
348 citations
TL;DR: The results show that the proposed models to predict bus arrival times at the same bus stop but with different routes are more accurate than the models based on the bus running times of single route.
Abstract: Provision of accurate bus arrival information is vital to passengers for reducing their anxieties and waiting times at bus stop. This paper proposes models to predict bus arrival times at the same bus stop but with different routes. In the proposed models, bus running times of multiple routes are used for predicting the bus arrival time of each of these bus routes. Several methods, which include support vector machine (SVM), artificial neural network (ANN), k nearest neighbours algorithm (k-NN) and linear regression (LR), are adopted for the bus arrival time prediction. Observation surveys are conducted to collect bus running and arrival time data for validation of the proposed models. The results show that the proposed models are more accurate than the models based on the bus running times of single route. Moreover, it is found that the SVM model performs the best among the four proposed models for predicting the bus arrival times at bus stop with multiple routes.
297 citations