Identification and mapping of asphalt surface deterioration by tyre-pavement interaction noise measurement

This paper proposes a novel approach to detect the pavement surface deterioration using tyre/road noise produced from a sensorized test-vehicle during a regular driving. The acoustic data have shown to be relevant on automatic asphalt-damage classification. In this proof of concept, four classes of bituminous road-surface states were classified, showing very promising results. The application of this technique will allow automated inspection, geolocation and prioritization of roads or street areas that may need maintenance and repair, in addition to informing drivers or autonomous vehicles about road safety conditions in order to prevent accidents. With this approach any vehicle can easily become an asphalt status tester with the installation of a microphone, enabling the possibility to transform any fleet of vehicles into a networked signal acquisition system that may form part of the decision support of administrations in the management and maintenance of road infrastructures, at a very low cost.