Local Gaussian Processes for Efficient Fine-Grained Traffic Speed Prediction

Abstract

Abstract-Traffic speed is a key indicator for the efficiency of an urban transportation system. Accurate modeling of the spatiotemporally varying traffic speed thus plays a crucial role in urban planning and development. This paper addresses the problem of efficient fine-grained traffic speed prediction using big traffic data obtained from static sensors. Gaussian processes (GPs) have been previously used to model various traffic phenomena, including flow and speed. However, GPs do not scale with big traffic data due to their cubic time complexity. In this work, we address their efficiency issues by proposing local GPs to learn from and make predictions for correlated subsets of data. The main idea is to quickly group speed variables in both spatial and temporal dimensions into a finite number of clusters, so that future and unobserved traffic speed queries can be heuristically mapped to one of such clusters. A local GP corresponding to that cluster can then be trained on the fly to make predictions in real-time. We call this method localization. We use non-negative matrix factorization for localization and propose simple heuristics for cluster mapping. We additionally leverage on the expressiveness of GP kernel functions to model road network topology and incorporate side information. Extensive experiments using real-world traffic data collected in the two U.S. cities of Pittsburgh and Washington, D.C., show that our proposed local GPs significantly improve both runtime performances and prediction accuracies compared to the baseline global and local GPs.

A comprehensive study of speed prediction in transportation system: From vehicle to traffic

“…Besides, the kernel function is the core of a Gaussian process which measures the distance between two sample points. The probability interpretation and CIs of output can be provided by non-parametric Bayesian formulation ( Le et al., 2017 ) in Gaussian Process.…”

“…For example, the weather information and traffic properties were considered in prediction through Gaussian process ( Chen et al., 2014 ), and the social mediadata and car trajectory data were taken into account in the same way in ( Lin et al., 2018 ). However, the cubic learning computation and quadratic space requirement are the major limitations of the Gaussian process ( Le et al., 2017 ).…”

DistTune: Distributed Fine-Grained Adaptive Traffic Speed Prediction for Growing Transportation Networks

“…Several variants of the k-NN and RNN methods were then introduced for traffic prediction, such as Bustillos and Chiu ( 3 ) and Cui et al ( 21 ). Le et al ( 22 ) addressed traffic speed prediction using big traffic data obtained from static sensors and proposed local Gaussian processes to learn and make predictions for correlated subsets of data. Jiang and Fei ( 23 ) introduced a data-driven vehicle speed prediction method based on Hidden Markov models.…”

Short-Term Traffic Prediction with Vicinity Gaussian Process in the Presence of Missing Data

“…1 K times with different subsets. The expectationκ of the RMSE rank corresponding to the minimum distance κ i , i = 1, · · · , K is (17). Then η is determined by (18).…”

“…However, GP still suffer from cubic time complexity in the size of training data. Fortunately, both parallel/distributed computation [15] and non-negative matrix factorization (NMF) techniques [17] provide the possibilities to decrease the computational complexity.…”