Analisi e Modelli per il Monitoraggio del Manto Stradale

A deep knowledge of road surface conditions is essential for safe driving. Moreover, the road surface conditions are subject to change over time due to wear and tear, weather phenomena, accidental events, ground level structural failures and many other factors, not least the rebuilding activities by the management authority. The vision of a detailed and updated framework of the conditions of the road network is therefore an essential requirement in order to plan an optimal maintenance and cope with emergencies. Eventually, the knowledge of the road conditions allows car drivers and motorcyclists to drive more safely. The latest generation of mobile devices (smartphones) are equipped with several sensors such as triaxial accelerometers, compasses, and GPS receivers, that make it possible to extract the characteristics of the road surface after appropriate digital signal processing. In fact, the sensitivity and the sampling frequency of the embedded accelerometers (about 100 Hz, recently elevated – in some models – up to 300 Hz) are adequate to detect and measure all vibrations, perceived by car drivers, which are induced by the road surface condition. In the SmartRoadSense research project, a mobile device is used as a navigation system and is rigid anchored in the front passenger area. In this way, its accelerometer is used to monitor the road surface roughness, causing vibrations not attributable to the operation of the engine itself. The analysis of these sampled signals allows to carry out an accurate classification of the type of pavement, recognizing asphalt injuries or failures. The dependence of these assessments on vehicle speed are properly compensated thanks to the information provided by the satellite navigation system. Furthermore, the computing and communication resources of mobile devices allow the real-time processing and transmission of data and other geo-referenced information to a central server. The server aggregates data provided by different mobile devices, which operate in the area, and makes it available on road maps – for ease of viewing – for the benefit of the managing authorities and motorists. Data reliability is guaranteed by the accuracy of the instruments and by the comparison of the measurements provided by the multiple users that have traveled the same road path. The cooperation with public transport companies and the installation of this technology on public transports, including busses, makes monitoring even more reliable and systematic. The SmartRoadSense project proposes the creation of a road collaborative system for automatic detection of road conditions based on the combined use of triaxial accelerometers and GPS receivers that are embedded in modern mobile devices. Notably, the project – in its entirety – can be summarized in three big main blocks: 1. The development of mathematical models based on predictive algorithms. These models – by numerical signal processing of the data provided by the accelerometer and the speed provided by the GPS sensor – allow the automatic classification of road conditions, without any user intervention. 2. The development of an application (currently freely distributed online for Android OS) that sends the road roughness georeferenced data to a central server. 3. The development of a server-side application that receives data sent from various mobile devices and aggregates them by performing a weighted average over time and space. This application generates an open database, posing particular attention to the users’ privacy (with a zero-knowledge approach), which allows end-users to extract and reuse the aggregated data. The research project was conducted at the Department of Pure and Applied Sciences (DiSPeA) of the University of Urbino involving many academic researchers with several skills: computer science, mathematics, electronics and geophysics. This thesis presents the road surface analysis performed in the SmartRoadSense project, the theoretical models, and the analysis on data obtained within the project in its various phases.