Opportunistic in-vehicle noise measurements assess road surface quality to improve noise mapping: Preliminary results from the MobiSense project

Abstract

The quality of road pavements affects noise emission caused by tire-road interactions. This in turn affects the health and well-being of residents near these roads. Road pavement quality degrades over time due to wear, accidents, and infrastructure works. These local features are usually not included in noise mapping due to the lack of high-quality information on pavements with enough spatial resolution. The aim of Mobisense is to assess the quality of the road surface by performing opportunistic noise and vibration measurements inside vehicles that are on the road for other purposes than road quality measurement. In the demonstrator phase of the project, 20 vehicles collect data while the drivers make their usual trips. Measurements from all vehicles are combined using machine learning techniques. This removes engine noise, corrects for vehicle specific speed dependence, and finally determines a rolling noise proxy in third-octave bands. This rolling noise correction includes the effect of pavement type as well as the effect of road surface degradation. This local variation in road surface quality is included as a correction in the rolling noise component of CNOSSOS and used to calculate a subset of the noise map for the Flemish region in Belgium. Including road surface quality in this way changes noise maps locally over a range of 6 dBA.