Here Global B.v.

Deepmap Inc.

Nvidia Corporation

Leica Geosystems Ag

Jeffrey Minoru Adachi

Antonio Haro

Jane MacFarlane

Mark Damon Wheeler

Xiaoqing Liu

Chen Chen

Jonathan Apollo Kung

Richard William Bukowski

Aaron J Dannenbring

Guangli Wu

Laura Michele Downs

Gregory William Coombe

Derik Schroeter

Andrew Lewis

Victor Lu

Ian Endres

Greg Coombe

A vehicle, for example, an autonomous vehicle receives signals from a global navigation satellite system (GNSS) and determines accurate location of the vehicle using the GNSS signal. The vehicle performs localization to determine the location of the vehicle as it drives. The autonomous vehicle uses sensor data and a high definition map to determine an accurate location of the autonomous vehicle. The autonomous vehicle uses accurate location of the vehicle to determine RTK corrections that is used for improving GNSS location estimates at a future location. The RTK corrections may be transmitted to other vehicles.

Accuracy of global navigation satellite system based positioning using high definition map based localization

A system align point clouds obtained by sensors of a vehicle using kinematic iterative closest point with integrated motions estimates. The system receives lidar scans from a lidar mounted on the vehicle. The system derives point clouds from the lidar scan data. The system iteratively determines velocity parameters that minimize an aggregate measure of distance between corresponding points of the plurality of pairs of points. The system iteratively improves the velocity parameters. The system uses the velocity parameters for various purposes including for building high definition maps used for navigating the vehicle.

Iterative closest point process based on lidar with integrated motion estimation for high definition maps

Drone space is defined according to a building model and a buffer space. At least one three-dimensional geometry is identified from the building model. The buffer space is calculated from the three-dimensional geometry. Coordinates for a drone air space are defined based on the buffer space. At least one path segment may be identified based on the coordinates for the drone air space, and the coordinates for drone air space are stored in a map database in association with the at least one path segment.

Air space maps

An approach involves receiving pixel-level labeling data for an image depicting at least a portion of a building facade. The pixel-level labeling data labels each of a plurality of pixels of the image as either window pixels or non-window pixels. The approach further involves generating a window pattern based on window size data, window spacing data, or a combination thereof extracted from the pixel-level labeling data. The approach further involves computing a confidence score for the window pattern based on at least one observed value of at least one characteristic of the window pattern or a deviation of the observed value from at least one expected value, and generating an abstract texture based on the window pattern, the confidence score, and/or other probabilistic metrics.

Method and apparatus for generating an abstract texture for a building facade or model

An approach is provided for identifying objects present in mesh representation of a geo-location, generating accurate 3D models for the objects, and aligning the 3D models to their corresponding objects in an application. The approach comprises processing and/or facilitating a processing of textured three-dimensional mesh data in one or more regions of interest to cause, at least in part, a generation of at least one two-dimensional depth image representation. The approach further comprises causing, at least in part, a filtering of the textured three-dimensional mesh data in the one or more regions of interest to remove mesh data below at least one threshold height based, at least in part, on the at least one two-dimensional depth image representation. Additionally, the approach comprises processing and/or facilitating a processing of the filtered textured three-dimensional mesh data to cause, at least in part, a generation of at least one partial three-dimensional model, including one or more upper facades above the at least one threshold height, of one or more objects located within the one or more regions of interest.

Method and apparatus for identifying buildings in textured 3D mesh data and generating 3D building models

A high-definition map system receives sensor data from vehicles travelling along routes and combines the data to generate a high definition map for use in driving vehicles, for example, for guiding autonomous vehicles. A pose graph is built from the collected data, each pose representing location and orientation of a vehicle. The pose graph is optimized to minimize constraints between poses. Points associated with surface are assigned a confidence measure determined using a measure of hardness/softness of the surface. A machine-learning-based result filter detects bad alignment results and prevents them from being entered in the subsequent global pose optimization. The alignment framework is parallelizable for execution using a parallel/distributed architecture. Alignment hot spots are detected for further verification and improvement. The system supports incremental updates, thereby allowing refinements of sub-graphs for incrementally improving the high-definition map for keeping it up to date.

Alignment of data captured by autonomous vehicles to generate high definition maps

An approach is provided for generating a cleaned object model to represent an object in a mapping database. The approach includes receiving point cloud data depicting the object. The approach also includes processing the point cloud data to determine one or more surface points of the point cloud data. The one or more surface points represent one or more surfaces of the object. The approach further includes cutting a model of the object into one or more fragments using the one or more surface points. The one or more fragments include one or more object fragments and one or more non-object fragments. The approach further includes designating the one or more object fragments as the cleaned object model to represent the object in the mapping database.

Method and apparatus for generating a cleaned object model for an object in a mapping database

A method comprises obtaining data points, each comprised of a time stamp and measurement; dividing the data points into sequences of consecutive data points; limiting the maximum time between consecutive data points in the same sequence; limiting the maximum time between the earliest and latest data points in each sequence; calculating a polynomial of lowest transmission cost for each sequence; limiting the approximation error between the data points in a sequence and the associated polynomial; and transmitting, to a server, data based on the calculated polynomial.

Method for compressing and reconstructing data sampled from continuous functions

A method comprising receiving probe data indicative of a set of navigational signal measurements that is matched to a link segment, determining a set of measurement errors such that each measurement error of the set of measurement errors is a difference between a location indicated by the link segment and a location indicated by a navigational signal measurement of the set of navigational signal measurements, determining at least one statistical attribute of the set of measurement errors, and storing an indication of the statistical attribute in map information associated with the link segment is disclosed.

Determination of a statistical attribute of a set of measurement errors

An approach is provided for classifying objects that are present at a geo-location and providing an uncluttered presentation of images of some of the objects in an application such as a map application. The approach includes determining one or more regions of interest associated with at least one geo-location, wherein the one or more regions of interest are at least one textured three-dimensional representation of one or more objects that may be present at the at least one geo-location. The approach also includes processing and/or facilitating a processing of the at least one textured three-dimensional representation to determine at least one two-dimensional footprint and three-dimensional geometry information for the one or more objects. The approach further includes causing, at least in part, a generation of at least one two-dimensional image representation of the one or more regions of interest by causing, at least in part, a projection of three-dimensional texture information of the at least one textured three-dimensional representation onto the at least one two-dimensional footprint. The approach also includes causing, at least in part, a classification of the one or more objects based, at least in part, on the at least one two-dimensional image representation and the three-dimensional geometry information.

Method and apparatus for classifying objects and clutter removal of some three-dimensional images of the objects in a presentation

An approach is provided for determining new point(s)-of-interest based, at least in part, on mobile device positioning, three dimensional location data, or a combination thereof. The approach involves processing and/or facilitating a processing of location information associated with a plurality of devices to determine one or more location points at which there are one or more concentrations of the plurality of devices. The approach also involves causing, at least in part, an accessing of three-dimensional data representing the one or more location points. The approach further involves processing and/or facilitating a processing of the three-dimensional data to determine one or more features that are indicative of a presence of one or more points of interest. The approach also involves determining one or more candidate points of interest based, at least in part, on the one or more features.

Method and apparatus for providing point-of-interest detection via feature analysis and mobile device position information

A computer model of a physical structure (or object) can be generated using context-based hypothesis testing. For a set of point data, a user selects a context specifying a geometric category corresponding to the structure shape. The user specifies at least one seed point from the set that lies on a surface of the structure of interest. Using the context and point data, the system loads points in a region near the seed point(s), and determines the dimensions and orientation of an initial surface component in the context that corresponds to those points. If the selected component is supported by the points, that component can be added to a computer model of the surface. The system can repeatedly find points near a possible extension of the surface model, using the context and current surface component(s) to generate hypotheses for extending the surface model to these points. Well-supported components can be added to the surface model until the surface of the structure of interest has been modeled as far as is well-supported by the point data.

Growing community of inventors

El Cerrito, CA, United States of America

Jeffrey Minoru Adachi