Road asset detection model using smartphones: A deployment analysis

Road assets, such as road signs, street names and other objects like fire hydrants and utility poles are objects closely monitored by government bodies and transportation authorities. More recently, autonomous driving requires road asset information as well. This paper presents a collection model to efficiently gather the information needed for road asset management and other digital cartography applications. Instead of using dedicated surveying vehicles operated by skilled drivers and high-quality sensors that are typical for traditional data collection performed by large mapping companies, it envisions an infrastructure in which the data collected through much more affordable sensors that mainstream consumer cars can host. An Android application named vidMAp based on a smartphone was developed that can be installed on smartphones in public vehicles, to collect information from the road scene. The presented model is analysed from a cost perspective as a function of the number of host vehicles and detection rate.