Bayerische Motoren Werke Aktiengesellschaft

Qianshan Li

Wanli Jiang

Maximilian Doemling

Sebastian Granzow

Tao Xu

Tomas Xu

Bill Lv

Jianpeng Li

Wenxin Tian

Shuhan Lv

Hongshan Xu

Rick Li

Ivan Xu

Carsten Isert

Gavin Xu

Gary Lu

Philipp Friedrich

Jenny Yuan

A method and apparatus evaluates a map for autonomous driving. The method acquires pixel points of lane lines of a road in a static map; acquires pixel points of lane lines of the road in a real-time map; forms first fitted curves of the lane lines in the static map by fitting the acquired pixel points of the lane lines in the static map respectively; forms second fitted curves of the lane lines in the real-time map by fitting the acquired pixel points of the lane lines in the real-time map respectively; calculates fitted-curve-similarities between the first fitted curves and the corresponding second fitted curves; and evaluates the consistency of the static map and the real-time map based on the fitted-curve-similarities.

Method and apparatus for evaluating maps for autonomous driving and vehicle

A method includes obtaining a pre-defined roadmodel 2-dimensional (2D) manifold in 6D space, wherein the roadmodel 2D manifold in 6D space is pre-defined by transforming objects in a roadmodel from 3D space into 6D space, obtaining a current position of the vehicle and a target position of the vehicle on the horizontal plane; obtaining a piece of the roadmodel 2D manifold covering the current position and the target position; flattening the piece of the roadmodel represented by the piece of the roadmodel 2D manifold onto the horizontal plane while maintaining a distance between any two points from 3D space to the horizontal plane; and planning a trajectory for the vehicle from the current position to the target position in the flattened roadmodel on the horizontal plane.

Roadmodel manifold for 2D trajectory planner

A method for determining a 6-dimensional (6D) delta pose of a vehicle includes obtaining a pre-defined roadmodel 2-dimensional (2D) manifold in 6D space, wherein the roadmodel 2D manifold in 6D space is pre-defined by transforming objects in a roadmodel from 3D (x, y, z) space into 6D (x, y, z, r, p, y) space, wherein x and y represent positions of objects on a horizontal plane, and z, r, p and y represent the height, roll angle, pitch angle and yaw angle of the objects in real world; obtaining a delta 3D odometry (Δx, Δy, θ) from an odometer of the vehicle, wherein Δx and Δy represent the movements in the lateral and forward-reverse directions, and θ represents a current heading of the vehicle; projecting the obtained delta 3D odometry onto the roadmodel 2D manifold; and determining the 6D delta pose of the vehicle corresponding to the delta 3D odometry.

3D odometry in 6D space with roadmodel 2D manifold

A method and an apparatus identifies an object. The method includes at a first time instant, acquiring raw data output by a camera installed on a vehicle, the output raw data comprises one of a continuous data output format for a continuous model or a discrete data output format for a discrete model; feeding the raw data to a discrete statistic model to form discrete statistic model format data for the first time instant, the discrete statistic model format data includes a continuous part having the continuous data output format and a discrete part having the discrete data output format; fusing the discrete statistic model format data for the first time instant with historical discrete statistic model format data to update the discrete statistic model format data for the first time instant; and identifying the object from the updated discrete statistic model format data for the first time instant, wherein the object includes a lanemarking characterized by a curve.

Method and system for identifying object

A method and an apparatus optimizes scan data obtained by sensors on vehicle, and corrects trajectory for a vehicle/robot based on the optimized scan data. The method for optimizing the scan data obtained by scanning environment elements, includes: step of obtaining the scan data, including obtaining at least two frames of scan data respectively corresponding to different timings; step of cluster processing, based on the characteristic of the data points, including classifying the plurality of data points in each frame of the scan data into one or more clusters; step of establishing correspondence, among the at least two frames of scan data, including searching and obtaining at least one set of clusters having correspondence; step of optimizing clusters, among the at least two frames of scan data, including conducting calculation to each set of the at least one set of clusters having correspondence, to obtain optimized clusters respectively corresponding to each set of the at least one set of clusters having correspondence; and step of optimizing the scan data, including accumulating all optimized clusters to obtain an optimized scan date for the at least two frames of scan data.

Method and apparatus for optimizing scan data and method and apparatus for correcting trajectory

A method for segmentation of a point cloud includes receiving a first frame of point cloud from a sensor; segmenting the first frame of point cloud to obtain a first set of point clusters representing a segmentation result for the first frame of point cloud; receiving a second frame of point cloud from the sensor; mapping the first set of point clusters to the second frame of point cloud; determining points within the second frame of point cloud which do not belong to the mapped first set of point clusters; segmenting the points within the second frame of point cloud which do not belong to the mapped first set of point clusters to obtain a second set of point clusters; and generating a segmentation result for the second frame of point cloud by combining the first set of point clusters and the second set of point clusters.

Incremental segmentation of point cloud

A method for determining driving strategy of a vehicle includes obtaining localization information of the vehicle, a driving route of the vehicle, and a current driving velocity of the vehicle. The method further includes detecting and interpreting a current status of a traffic light. The method further includes recognizing a duration of the current status of the traffic light. The method further includes detecting a relevant stop line. The method further includes obtaining a distance between the stop line and the vehicle. The method further includes determining whether to change the current driving velocity according to the current status of the traffic light, the duration of the current status of the traffic light, the distance between the traffic stop line and the vehicle, and the current driving velocity. A data processing device for determining driving strategy of a vehicle is also provided.

Method and apparatus for determining driving strategy of a vehicle

A method and apparatus for selecting a map from a plurality of maps having different resolutions for a moving object includes dynamically selecting an appropriate map from maps having different levels of details according to environment information, such that a minimum amount of required map is loaded, thereby effectively improving the data processing efficiency of vehicles.

Methods and apparatus for selecting a map for a moving object, system, and vehicle/robot

A computer-implemented method for representing environmental elements includes receiving scan data comprising at least a point cloud representing at least an environmental element from a sensor, segmenting the point cloud into point clusters, and partitioning the point clusters into hierarchical grids. The method also includes establishing a Gaussian distribution for points in each cell of each of the hierarchical grids, and constructing a Gaussian Mixture Model based on the Gaussian distribution for representing the environmental element.

Method and apparatus for representing environmental elements, system, and vehicle/robot

A method for configuring a sensor, which is adapted to be installed on a moving object, includes obtaining localization information of the sensor from a localization device, obtaining characteristic information and localization information of at least one element from a digital map, and choosing one or more elements of the at least one element in the digital map. The method also includes configuring the sensor according to the characteristic information and/or the localization information of the one or more chosen elements.

Method, device and system for configuration of a sensor on a moving object

A method and an apparatus for uncertainty modeling of a point cloud and a non-transient storage medium are provided. The method finds the neighboring points for each point; fits a local quadratic surface for each point; and generates a Gaussian Model (GM) for representing the uncertainty modeling of the local surface at each point. The GM is constructed in a piece-wise manner on the underlying surface of the point cloud, which will accurately model the surface uncertainty.

Method and apparatus for uncertainty modeling of point cloud

A method and an apparatus are provided for representing a map element and for locating a vehicle/robot based thereupon. The method for representing a map element includes: generating a Gaussian Mixture Model for the map element; generating a signature for identifying the map element, wherein the signature includes properties of the map element; and generating a Signatured Gaussian Mixture Model for representing the map element, wherein the Signatured Gaussian Mixture Model includes the Gaussian Mixture Model and the signature.

Growing community of inventors

Shanghai, China

Qianshan Li