An initial attempt to develop an empirical relation between texture and pavement friction using the HHT approach

Pavement friction, one of the main factors contributing to road safety, depends mainly on surface texture. However, despite its importance having been well corroborated by the numerous investigations attempting to predict it, the manner in which texture is related to friction remains widely unknown. The current paper will explore the friction-texture relationship based on a new signal processing method called Huang-Hilbert Transformation, or HHT. This method allows empirical decomposition of a texture profile to a set of basic profiles in a limited number, called Intrinsic Mode Functions, or IMFs. Each IMF contains a given interval of amplitudes and frequencies. From the obtained IMFs, a set of four new functions called Base Intrinsic Mode Functions, or BIMF are computed based on the frequency and power content of the underlying IMFs and are characterized using the Hilbert Transformation technique to obtain the scale-dependent norm frequency and amplitude profiles. Then these two parameters are correlated with the pavement friction from a multiple regression analysis. This analysis is applied to a set of texture and friction data measured through test track surfaces in France and lab samples of concrete in the United States. The textures and frictions are measured with the Circular Texture Meter (CTMeter) and the Dynamic Friction Tester (DFTester), respectively. The obtained results show a good correlation between the BIMF's parameters to friction, thus opening a promising new means for characterizing texture in relation to friction.