Methodology for correcting erroneous individual vehicle speed data using locally weighted regression

Abstract

Effective detection and correction of outliers in raw traffic data collected from the field is of keen interest because reliable traffic information is highly dependent on the quality of raw data. Global positioning systems (GPS) based traffic surveillance systems are capable of producing individual vehicle speeds that are invaluable for various traffic management and information strategies. This study proposes a locally weighted regression (LWR) based filtering method for individual vehicle speed data. Both a technique to generate synthetic outliers and to injecting synthetic outliers are also presented to systematically evaluate the proposed method. A method to determine parameters associated with the LWR that affect the smoothing performance is devised and applied to obtain more reliable data correction. A set of illustrative evaluation examples explains that the proposed method is useful in filtering individual vehicle speed data.