The present paper proposes a solution for enhancing the environment representation of the world, in intersections. The proposed solution is to augment the ego-vehicle environment representation, currently limited to the information provided by the Sensorial Perception system with additional information from a proposed Extended Digital Map. A prior necessary step, the accurate ego-vehicle global localization, is achieved through an original approach of aligning the sensorial landmarks detected by the Sensorial On-board Perception system with the corresponding, accurately positioned, map landmarks existing in the Extended Digital Map. Prior to the data alignment, the ego-vehicle driving lane identification is required; a probabilistic approach in the form of a Bayesian Network performs the ego-vehicle driving lane identification based on the information from the two input systems. The proposed solution is adequate for use in an advanced driving assistance system for improving the travel and traffic information of the ego-vehicle with respect to an approaching intersection.
Improved localization and enhanced environment representation by sensorial and digital map data fusion in intersections
Verbesserte Lokalisierung und genauere Umfelddarstellung durch Sensor gestützte und digitale Kartendatenfusion von Straßenkreuzungen
2010
6 Seiten, 5 Bilder, 14 Tabellen, 2 Quellen
(nicht paginiert)
Conference paper
Storage medium
English
Data fusion for traffic flow estimation at intersections
German Aerospace Center (DLR) | 2011
|FAST AND PRECISE LOCALIZATION AT STOP INTERSECTIONS
British Library Conference Proceedings | 2013
|Fast and precise localization at stop intersections
IEEE | 2013
|