Abstract: A method is provided to using a machine learning model to predict lane geometry where incorrect or missing lane line geometry is detected. Methods may include: receiving a representation of lane line geometry for one or more roads of a road network; identifying an area within the representation as broken lane line geometry of an intersection using a machine learning model; generating a masked area of the broken lane line geometry of the intersection within the representation; processing the representation with the masked area using an inpainting model to generate an inpainted result within the masked area of restored lane line geometry of the intersection, where the inpainting model is trained using a set of representations identified as lane line geometry of intersections; and updating a map database to include the restored lane line geometry of the intersection in place of the broken lane line geometry of the intersection.

Abstract: A method, apparatus and computer program product are provided to process probe data in accordance with a map-centric map matching technique. In the context of a method, a plurality of vertices are defined along a polyline representative of a road segment such that the polyline includes one or more edges. Each edge extends between a pair of neighboring vertices. For each vertex of a respective edge, spatial searches are conducted to identify each probe point within a region about a respective vertex. For each probe point identified within a region about a respective vertex, a determination is made as to whether the probe point satisfies a projection criteria in order for the probe point to be projected onto the edge or one of the neighboring vertices between which the edge extends.

Abstract: A method, apparatus and computer program product are provided to determine the stacking order of two or more roads that are vertically aligned with one another. In the context of a method, error statistics associated with probe points representative of travel along a first road are determined. In an instance in which the first road is vertically aligned with a second road, the method also includes determining whether one of the first or second roads passes beneath the other of the first or second roads based at least in part upon the error statistics associated with the probe points representative of travel along the first road. A corresponding apparatus and computer program product are also provided.

Abstract: A way to represent roads as data is disclosed. Vehicle systems that provide driver assistance features relating to road curvature, such as curve warning systems or lane departure systems, use data that indicates road shape and data that indicate road position. To support these vehicle systems, a dual data representation of road geometry is provided. One road geometry data representation indicates the road position, e.g., road centerline or lane centerline, and the other road geometry data representation indicates the road shape, e.g., curvature and/or heading. Several different dual shape representations may be provided and can co-exist with existing and future database representations.