GPS Data-based Non-parametric Regression for Predicting Travel Times in Urban Traffic Networks

Marković, Hrvoje; Bašić, Bojana Dalbelo; Gold, Hrvoje; Dong, Fangyan; Hirota, Kaoru

doi:10.7307/ptt.v22i1.159

Cited by 16 publications

(15 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For search step length (d), [8] suggests it to be bigger than 2D + 1 according to Takens theorem, where D is the number of features. However, some research [17], [18] shows opposite results. Though v is also an important parameter, only a few studies considered it.…”

Section: B Parameters Strategiesmentioning

confidence: 91%

See 1 more Smart Citation

An Overview of Parameter and Data Strategies for k-Nearest Neighbours Based Short-Term Traffic Prediction

Sun

Cheng

Goswami

et al. 2017

Proceedings of the 2017 International Conference on E-Society, E-Education and E-Technology

View full text Add to dashboard Cite

Modern intelligent transportation systems (ITS) requires reliable and accurate short-term traffic prediction. One widely used method to predict traffic is k-nearest neighbours (kNN). Though many studies have tried to improve kNN with parameter strategies and data strategies, there is no comprehensive analysis of those strategies. This paper aims to analyse kNN strategies and guide future work to select the right strategy to improve prediction accuracy. Firstly, we examine the relations among three kNN parameters, which are: number of nearest neighbours (k), search step length (d) and window size (v). We also analysed predict step ahead (m) which is not a parameter but a user requirement and configuration. The analyses indicate that the relations among parameters are compound especially when traffic flow states are considered. The results show that strategy of using v leads to outstanding accuracy improvement. Later, we compare different data strategies such as flow-aware and time-aware ones together with ensemble strategies. The experiments show that the flowaware strategy performs better than the time-aware one. Thus, we suggest considering all parameter strategies simultaneously as ensemble strategies especially by including v in flow-aware strategies.

show abstract

Section: B Parameters Strategiesmentioning

confidence: 91%

“…Though v is also an important parameter, only a few studies considered it. Using v was not showing promising improvement in [12], the reason can be that they determined v separately from k and d. In some studies, both k and d get optimised [13], [17]. [8] shows we should optimise k and d at the same time.…”

Section: B Parameters Strategiesmentioning

confidence: 98%

An Overview of Parameter and Data Strategies for k-Nearest Neighbours Based Short-Term Traffic Prediction

Sun

Cheng

Goswami

et al. 2017

Proceedings of the 2017 International Conference on E-Society, E-Education and E-Technology

View full text Add to dashboard Cite

show abstract

“…Significant research effort has been focusing on models for predicting speed, travel times, and departure/arrival times on both spatial and temporal domains for different transportation modes. A variety of prediction methods, such as the Box–Jenkins models , the KF models , regression models , artificial neural networks , and microscopic simulation models were developed for various applications. A wide range of factors are considered in these studies, including dynamic traffic characteristics (e.g., volume, speed, vehicle composition, and driving behavior) and external factors (e.g., incidents, traffic control devices, roadway geometry, and weather) that affect daily traffic operation.…”

Section: Literature Reviewmentioning

confidence: 99%

An adaptive model for highway travel time prediction

Liu

Chien

Chen

2012

J of Advced Transportation

View full text Add to dashboard Cite

Traffic congestion caused by either insufficient road capacity or unexpected events has been a major problem in urban transportation networks. To disseminate accurate traveler information and reduce congestion impact, it is desirable to develop an adaptive model to predict travel time. The proposed model is practically implementable to capture dynamic traffic patterns under various conditions, which integrates the features of exponential smoothing and the Kalman filter by utilizing both real-time and historic data. The model is simple in formulation while robust in performance in terms of accuracy and stability. With a constraint or nonconstraint smoothing factor, the proposed model is tested with both real world and simulated data and demonstrated itself a sound model that outperforms others (e.g., Kalman filter and simple exponential smoothing) specifically under recurring and nonrecurring congestion. MARE, mean absolute relative error; SDAPE, standard deviation of absolute relative error; SES, simple exponential smoothing; ESES, enhanced SES; ESES-A, alternative ESES. 650 X. LIU ET AL.

show abstract

“…There have been many research efforts that are inspired by the need of short‐term travel time prediction. A wide spectrum of methodologies have been attempted, including traffic flow based models (Zhang and Rice, 2003; Vanajakshi et al, 2009), traffic simulation‐based approaches (Ruiz Juri et al, 2007), time‐series models (D’Angelo et al, 1999), various statistical and regression techniques (Markovic et al, 2010; Bhaskar et al, 2011), and Kalman filtering (Park and Rilett, 1998). Among these techniques, a data‐driven approach, the artificial neural network (ANN), has emerged as a robust tool and has gained significant research attention in a vast area of transportation applications (Dougherty, 1995; Adeli, 2001; Huisken, 2006).…”

Section: Introductionmentioning

confidence: 99%

Development of Recurrent Neural Network Considering Temporal‐Spatial Input Dynamics for Freeway Travel Time Modeling

Zeng

Zhang

2013

Computer aided Civil Eng

View full text Add to dashboard Cite

The artificial neural network (ANN) is one advance approach to freeway travel time prediction. Various studies using different inputs have come to no consensus on the effects of input selections. In addition, very little discussion has been made on the temporal–spatial aspect of the ANN travel time prediction process. In this study, we employ an ANN ensemble technique to analyze the effects of various input settings on the ANN prediction performances. Volume, occupancy, and speed are used as inputs to predict travel times. The predictions are then compared against the travel times collected from the toll collection system in Houston. The results show speed or occupancy measured at the segment of interest may be used as sole input to produce acceptable predictions, but all three variables together tend to yield the best prediction results. The inclusion of inputs from both upstream and downstream segments is statistically better than using only the inputs from current segment. It also appears that the magnitude of prevailing segment travel time can be used as a guideline to set up temporal input delays for better prediction accuracies. The evaluation of spatiotemporal input interactions reveals that past information on downstream and current segments is useful in improving prediction accuracy whereas past inputs from the upstream location do not provide as much constructive information. Finally, a variant of the state‐space model (SSNN), namely time‐delayed state‐space neural network (TDSSNN), is proposed and compared against other popular ANN models. The comparison shows that the TDSSNN outperforms other networks and remains very comparable with the SSNN. Future research is needed to analyze TDSSNN's ability in corridor prediction settings.

show abstract

GPS Data-based Non-parametric Regression for Predicting Travel Times in Urban Traffic Networks

Abstract: A model for predicting travel times by mining spatiotemporal data acquired from vehicles equipped with GlobalPositioning System (GPS)

Cited by 16 publications

References 27 publications

An Overview of Parameter and Data Strategies for k-Nearest Neighbours Based Short-Term Traffic Prediction

An Overview of Parameter and Data Strategies for k-Nearest Neighbours Based Short-Term Traffic Prediction

An adaptive model for highway travel time prediction

Development of Recurrent Neural Network Considering Temporal‐Spatial Input Dynamics for Freeway Travel Time Modeling

Contact Info

Product

Resources

About