Data Fusion Based Hybrid Approach for the Estimation of Urban Arterial Travel Time

Abstract

Travel time estimation in urban arterials is challenging compared to freeways and multilane highways. This becomes more complex under Indian conditions due to the additional issues related to heterogeneity, lack of lane discipline, and difficulties in data availability. The fact that most of the urban arterials in India do not employ automatic detectors demands the need for an effective, yet less data intensive way of estimating travel time. An attempt has been made in this direction to estimate total travel time in an urban road stretch using the location based flow data and sparse travel time data obtained using GPS equipped probe vehicles. Three approaches are presented and compared in this study: (1) a combination of input-output analysis for mid-blocks and Highway Capacity Manual (HCM) based delay calculation at signals named as base method, (2) data fusion approach which employs Kalman filtering technique (nonhybrid method), and (3) a hybrid data fusion HCM (hybrid DF-HCM) method. Data collected from a stretch of roadway in Chennai, India was used for the corroboration. Simulated data were also used for further validation. The results showed that when data quality is assured (simulated data) the base method performs better. However, in real field situations, hybrid DF-HCM method outperformed the other methods.

SVM‐based hybrid approach for corridor‐level travel‐time estimation

“…Lower scores for MAPE values indicate that the estimation accuracy of the model is good. Many studies have used the MAPE values as the measure of accuracy [2, 43, 69]. The formula used for calculating the MAPE and MAD values are given below: (a) MAPE where is the observed value of travel‐time (s), is the predicted value of travel‐time (s), n number of observations. MAD: The MAD of a data set is the average distance between each data point and the mean.…”

Addressing errors in automated sensor data for real‐time traffic state estimation using dynamical systems approach

“…Additionally, most of the above studies dealt with estimation of spatial variables from location‐based data only, which may miss out the spatial variations making them less effective. Studies reported by Anand et al [19] and Anusha et al [29] have investigated the use of data from multiple sources, generally termed as data fusion, in order to deal with limitations of using location‐based data from a single source in estimation problems under heterogeneous and less lane disciplined traffic conditions. However, both these studies used manually collected data (which would typically be more accurate) for the corroboration of the estimation scheme.…”

Analysed potential of big data and supervised machine learning techniques in effectively forecasting travel times from fused data

“…Yu et al use support vector machines (SVM) to predict bus arrival times at bus stations [5], Zong et al apply a genetic algorithm to forecast daily commute travel times in Beijing [13] and Simroth uses a nonparametric distribution-free regression model [14]. Regarding the data sources used for the travel time forecasting these are mainly global positioning system (GPS) based data [14,15,16], survey data [13] or data from different types of road detectors [6,7]. Multiple data sources are rarely used [5].…”