On the recursive estimation of vehicular speed using data from a single inductance loop detector: a Bayesian approach

This paper investigates the recursive estimation of vehicular speed using the information provided by a single inductance loop detector (ILD). A statistical model for space-mean speed measured by an ILD is developed, upon which a Bayesian analysis is carried out to estimate vehicular speed. This results in a set of recursive formulae which is analytically nice and neat. The incurred computational cost for updating the estimate of vehicular speed is kept to be a minimum. As a by-product, a simple method for the calibration of the effective vehicle length of an ILD is also developed. The proposed method is illustrated using simulation studies and a practical example.