Autonomous robot navigation based on OpenStreetMap geodata

Abstract

This paper introduces the appliance of standardized, free to use and globally available geodata for autonomous robot navigation. For this, data from the famous collaborative OpenStreetMap (OSM) mapping project are used. These geodata are public domain and include rich information about streets, tracks, railways, waterways, points of interest, land use, building information and much more. Beyond the spatial information, the geodata contain detailed information about the name, type and width of the streets as well as public speed limits. As a contribution of this paper, the OSM data are integrated for the first time into the robot tasks of localization, path planning and autonomous vehicle control. Following the description of the approach, experimental results in outdoor environments demonstrate the effectiveness of this approach.