公开(公告)号：US20250148763A1

公开(公告)日：2025-05-08

申请号：US18936575

申请日：2024-11-04

Applicant: ROBERT BOSCH GmbH ,  CARIAD SE

Inventor： Jared Evans ,  Krishnan Bharath Navalpakkam ,  Giacomo Bassetto

IPC: G06V10/77 ,  G06V10/75 ,  G06V10/764 ,  G06V10/774 ,  G06V20/58

Abstract: The invention relates to a method (100) for dimension reduction of a multi-dimensional feature space for training a machine learning model (50) by machine learning, comprising the following steps: providing (101) at least one data pair (30), in which in each case an original data element (31) and a modified data element (32) have a feature difference (Δf) in relation to one another, which feature difference is specific to a respective defined task for machine learning, determining (102) at least one task-specific feature space, which is specific to the at least one feature difference (Δf), on the basis of a comparison of the respective data pairs (30), performing (103) the dimension reduction on the basis of the determined task-specific feature space.

公开(公告)号：US20250104454A1

公开(公告)日：2025-03-27

申请号：US18895832

申请日：2024-09-25

Applicant: CARIAD SE ,  ROBERT BOSCH GmbH

Inventor： Julio Borges ,  Praveen Anil Kulkarni ,  Attila Juhasz

IPC: G06V20/70 ,  G06T7/12 ,  G06T11/00 ,  G06V10/25 ,  G06V10/764 ,  G06V20/54

Abstract: The invention relates to a method (100) for providing synthetic data for use in machine learning, wherein the synthetic data comprise synthetic, photorealistic image information, said method comprising the following method steps: generating (101) simulation data, wherein the simulation data are in the form of annotated data and are generated by a simulation environment, determining (102) annotations from the generated simulation data, generating (103) the synthetic data by means of semantic image synthesis (SIS) using the determined annotations as input for the semantic image synthesis (SIS), annotating (104) the generated synthetic data based on an output from the simulation environment.

公开(公告)号：US20250091573A1

公开(公告)日：2025-03-20

申请号：US18885279

申请日：2024-09-13

Applicant: CARIAD SE ,  VOLKSWAGEN AKTIENGESELLSCHAFT

Inventor： Michael Stark ,  Daniel Münning

IPC: B60W30/12 ,  B60W40/02

Abstract: The disclosure relates to validation of a driving corridor for lateral guidance of a motor vehicle by a driver assistance system. Based on environmental data, a driving corridor with corridor boundaries is determined, the corridor boundaries specifying a first track width for the lateral guidance of the motor vehicle. It is then checked whether there is more than one driving trajectory within the corridor boundaries, by determining whether there is more than one data point along a rear axle of the motor vehicle, each of which is assigned to a different driving trajectory. Based on a result of the determining, the driving corridor is validated and the first track width is used for the lateral guidance of the motor vehicle, or the driving corridor is falsified and a different, second track width is used for the lateral guidance of the motor vehicle.

公开(公告)号：US12236722B2

公开(公告)日：2025-02-25

申请号：US17523882

申请日：2021-11-10

Applicant: CARIAD SE ,  AUDI AG

Inventor： David Mai ,  Edson Fangwa Nzeya ,  Sandeep Kumar Patil

IPC: G07C5/00 ,  G06F9/50 ,  G07C5/08 ,  H04W4/40

Abstract: Approaches, techniques, and mechanisms are disclosed for large scale vehicle data collaborative analysis. According to one embodiment, a large amount of data streams is received from a multitude of vehicles. A distributed event streaming system is applied to parse the data streams based on an attribute, such as time sensitive data, location-specific data, or vehicle maintenance, operational, or fault-prevention data. A data container platform instance hosts applications that receive the parsed data streams. Output of an application is transformed into data streams having a common topic. Other applications may access the data streams by topic. Upon receiving an indication that an application has processed a data stream by topic, a decision point may be reached and executed by an external application, triggering an action on a vehicle from the multitude of vehicles.

公开(公告)号：US20250061700A1

公开(公告)日：2025-02-20

申请号：US18806574

申请日：2024-08-15

Applicant: CARIAD SE ,  Robert Bosch GmbH

Inventor： Daniel BAUER ,  Sergej NEUMANN

IPC: G06V10/82 ,  G06V10/25 ,  G06V10/44 ,  G06V10/764 ,  G06V10/766

Abstract: The disclosure relates to determining an object structure of an object, such as lane edges and/or lane markings. Image data describing an image of an environment with the object located therein are provided and then fed into at least one neural network that is trained to determine feature data, including predetermined features with respect to geometric shapes and/or colors of the object. Edge endpoints of the object structure of the object are determined by applying an edge endpoint determiner, and then an ROI operation is performed on at least one surrounding area around at least one determined edge endpoint. At the same time, intermediate graph points of the object structure are determined by applying an intermediate graph point determiner. The final object structure is determined and/or marked based on the determined edge endpoints and intermediate graph points. The determined object structure is provided to a computer vision functionality.

公开(公告)号：US12153935B2

公开(公告)日：2024-11-26

申请号：US17989576

申请日：2022-11-17

Applicant: CARIAD SE

Inventor： Daniel Kerk

IPC: G06F1/00 ,  G06F1/3206 ,  G06F9/4401

Abstract: A method for operating a control unit of a motor vehicle, wherein a state management module of a runtime environment initiates a standby mode of the runtime environment as a function of a stop request. The invention provides that, as a function of the stop request, the state management module first sends out to at least one application a respective backup request which requests the respective application to store respective predetermined runtime data in a non-volatile data memory, and then, however, if a resume request is subsequently received within a waiting time period, the normal operating mode is continued without the planned switch-over to the standby mode.

公开(公告)号：US12145619B2

公开(公告)日：2024-11-19

申请号：US17860936

申请日：2022-07-08

Applicant: Cariad SE ,  Volkswagen Aktiengesellschaft

Inventor： Sven-Garrit Czarnian ,  Thomas Dammeier ,  Michael Holicki ,  Ralph Hänsel ,  Timo Iken ,  Roland Kube ,  Carolin Last ,  Stefan Wappler

IPC: B60W60/00 ,  B60W40/105 ,  G01S13/931 ,  G06V20/56 ,  G06V20/58

Abstract: According to a method for self-localization of a vehicle, a first pose of the vehicle is determined in a map coordinates system, based on environment sensor data representing an environment of the vehicle, a landmark is detected in the environment, a position of the landmark is determined in the map coordinates system and a second pose of the vehicle is determined in the map coordinates system dependent on the position of the landmark. An assignment instruction is consulted, matching up the first pose with at last one preferred sensor type or at least one dominant landmark type. Depending on the assignment instruction, a first environment sensor system is activated and a second environment sensor system is deactivated, whereupon the environment sensor data are generated by means of the first environment sensor system.

公开(公告)号：US20240303919A1

公开(公告)日：2024-09-12

申请号：US18598912

申请日：2024-03-07

Applicant: CARIAD SE ,  ROBERT BOSCH GmbH

Inventor： Denis Tananaev

IPC: G06T17/00 ,  G06T7/11

CPC classification number: G06T17/00 ,  G06T7/11 ,  G06T2207/20081 ,  G06T2207/20084

Abstract: The invention relates to a method (100) for generating a representation (70) of the surroundings, comprising the following steps:



providing (101) at least one image (30) that results from a recording by an image detection device (5) and that represents objects (6) and/or surfaces (6) in the surroundings (7) of the image detection device (5), wherein the provided image (30) is subdivided into multiple image columns (31),
generating (102) the representation (70) of the surroundings, wherein for this purpose multiple three-dimensional stixels (80) for each image column (31) of the provided image (30) are parameterized for representing the objects (6) and/or surfaces (6) in three-dimensional space,
wherein the generation (102) of the representation (70) of the surroundings takes place using a model (50) which uses the provided image (30) as input.

公开(公告)号：US20240300504A1

公开(公告)日：2024-09-12

申请号：US18597824

申请日：2024-03-06

Applicant: CARIAD SE ,  AUDI AG

Inventor： Bernd Schaefer ,  Andreas Sticht ,  Johann Knoeferl ,  Dominic Rothenwöhrer

IPC: B60W50/06 ,  B60W50/00

CPC classification number: B60W50/06 ,  B60W2050/0043 ,  B60W2050/0083 ,  B60W2510/0638 ,  B60W2556/10

Abstract: The disclosure relates to a method for calculating a setpoint value curve for an intervention in an operation of a vehicle component of a vehicle in order to set an operating variable to a target value in the vehicle component. The operating variable is cyclically compared with a start value, and if it is detected that the operating variable has the start value, a setpoint trajectory for setting the operating variable along a time curve in accordance with a predetermined curve shape to the target value starts by an intervention in a manipulated variable which acts on the operating variable.

公开(公告)号：US20240199077A1

公开(公告)日：2024-06-20

申请号：US18572153

申请日：2021-06-24

Applicant: CARIAD SE ,  Volkswagen Aktiengesellschaft

Inventor： Anna REHR ,  Simon GROSSJOHANN

IPC: B60W60/00

CPC classification number: B60W60/0011

Abstract: A motion planning system and method for motion planning of an autonomous vehicle are provided, wherein a motion planning system determines a plurality of configuration samples through a configuration space, wherein the configuration samples are determined in predefined sampling steps, wherein a sequence of configuration samples defines a possible trajectory of the vehicle in the configuration space, wherein the plurality of configuration samples in each sampling step, are determined in parallel by using at least one graphics processing unit of the motion planning system, wherein sequences of configuration samples are calculated from all configuration samples by predefined motion criteria, and wherein the calculated sequences of configuration samples are provided as vehicle trajectories for motion of the vehicle.