Abstract

In the future, vehicles will gather more and more spatial information about their environment, using on-board sensors such as cameras and laser scanners. Using this data, e.g. for localization, requires highly accurate maps with a higher level of detail than provided by today's maps. Producing those maps can only be realized economically if the information is obtained fully automatically. It is our goal to investigate the creation of intermediate level maps containing geo-referenced landmarks, which are suitable for the specific purpose of localization. To evaluate this approach, we acquired a dense laser scan of a 22 km scene, using a mobile mapping system. From this scan, we automatically extracted pole-like structures, such as street and traffic lights, which form our pole database. To assess the accuracy, ground truth was obtained for a selected inner-city junction by a terrestrial survey. In order to evaluate the usefulness of this database for localization purposes, we obtained a second scan, using a robotic vehicle equipped with an automotive-grade laser scanner. We extracted poles from this scan as well and employed a local pole matching algorithm to improve the vehicle's position. Document type: Part of book or chapter of book

Abstract

International audience; Traditionally, national mapping agencies produced datasets and map products for a low number of specified and internally consistent scales, i.e. at a common level of detail (LoD). With the advent of projects like OpenStreetMap, data users are increasingly confronted with the task of dealing with heterogeneously detailed and scaled geodata. Knowing the scale of geodata is very important for mapping processes such as for generalization of label placement or land-cover studies for instance. In the following chapter, we review and compare two concurrent approaches at automatically assigning scale to OSM objects. The first approach is based on a multi-criteria decision making model, with a rationalist approach for defining and parameterizing the respective criteria, yielding five broad LoD classes. The second approach attempts to identify a single metric from an analysis process, which is then used to interpolate a scale equivalence. Both approaches are combined and tested against well-known Corine data, resulting in an improvement of the scale inference process. The chapter closes with a presentation of the most pressing open problem