Développement d’un système pour le suivi du comportement fonctionnel et structural des chaussées

Abstract

This research proposes an intelligent acquisition model for monitoring the condition of roadways and traffic on them. The main objective is to create connected pavements using an infrastructure management platform, with the aim of offering pavement interventions at the lowest cost to society. The structure of the roadway can thus be permanently monitored, and the transfer of information is carried out in real time and continuously, with the support of cellular modules, ultimately 5G, allowing the quality of the structure to be qualified at all times. and from anywhere. The interest of such an innovation for all stakeholders in the road industry is significant, as much for the ministries as for the municipalities, consulting engineers, suppliers, academics and of course the population who must have access profits associated with the taxes it pays. As such, the primary goal of this research is to design and validate a new mechanical system inserted under the surface layer of the road, which measures the deformations and the settlement of it at different depths and at several horizontal spacings. This system measures electrical characteristics which are well correlated with deflection or compaction basins under load. From these deflection basins, the research thus highlights the possibility of determining the charges, via these sensors, and of sending the information collected in a database. The subsequent analysis then makes it possible to determine: • The compaction of the road layers; and, • The intensity of the loads on a given roadway. This analysis is carried out by correlating the settlement of the road layers, deduced from the electrical characteristics, and the load, via the stiffness modules. The system is integrated into the roadway in order to deduce the settlements, retaining only the significant dynamic distortions over time and thereby eliminating the noise. The system then transmits the data collected to the acquisition model which transforms it into information that can be used by road network managers, that is to say axle loads. The acquisition results make it possible to obtain the load spectrum that the pavement undergoes in real time, which represents an important input to the structural design and evaluation software for pavements AASHTO 2002. A prototype is designed in the laboratory of coated with ETS and the tests carried out in order to identify the function which links the electrical characteristics and the loads. This proposed module is operational. The second goal of this research is to design and validate an integrated LiDAR-video approach which automatically associates a level of degradation of a road surface with an image collected with a camera and a laser survey by LiDAR (Time of Flight). These digital and depth data make it possible to qualify, independently, with a supervised learning model, vertical deformations and surface cracks (potholes, longitudinal, transverse or mesh cracks, ruts, etc.). This digital and depth data also makes it possible to build a functional database to have an overview of the condition of the roadway, at a specific point in its life cycle. This then becomes a reliable reference base allowing to ultimately select the most appropriate maintenance or rehabilitation intervention according to the location, type and severity of the damage identified and categorized, and the diagnosis of their causes, while having an evolution of the profile of degradations over time (deformations and cracks).

Date	11 Oct 2020
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Gabriel J. Assaf (Supervisor)