Analyzing road quality using a mobile application

Abstract

This paper presents the current progress in acquiring and analyzing the data, obtained from smartphone sensor measurements during a drive with a car. The main aim of the research is to create generic rules to identify road state and its quality. Previously, the mobile application has been created, and measurements began. Currently, we acquired data from more than 500km of daily driving and created a road map using binning and basic clustering. The aggregated values on the map contain information about speed, location and peek linear 3-axial acceleration. In the future, we expect to apply Kalman filtering for cleansing the acceleration data and to continue with evaluating other quality metrics.