This paper presents two localization strategies for vehicles localization in urban areas when GPS data are unavailable or have a poor quality due to multi-paths, satellites bad visibility or masks. The presented methods rely on approaches that take profit of a new source of information, which is a virtual 3D city model. The virtual 3D model is used to simulate sensors data such as data from a camera or a 2D laser range finder. Then these generated virtual data are aligned with the real data acquired by the embedded sensors in order to estimate the location of the vehicle. Results obtained with real data acquired in real conditions prove the feasibility of such approaches.