Estimation of Individual Vehicle Speed Using Inductive Wave Forms

Inductive loop detectors (ILDs) have been widely used for obtaining invaluable real-time traffic parameters in the field. One of the research topics associated with ILDs is to extract travel speed information from single loop detectors, which is the major focus of this study. This study develops statistical models to estimate vehicle speeds using inductive wave forms that are extracted by the impedance analyzer. Two statistical models are developed to be used in different traffic conditions such as 'congested' and 'non-congested'. A nice feature of this study is that individual vehicle speed is obtainable, which can be further utilized to produce more accurate real-time traffic information. Promising estimation accuracies in terms of mean absolute percentage error (MAPE), 8.2% and 8.0% for non-congested and congested traffic conditions respectively, have demonstrated that the proposed method would be a useful tool for supporting traffic surveillance systems.