公开(公告)号：US20230245469A1

公开(公告)日：2023-08-03

申请号：US18160236

申请日：2023-01-26

Applicant: CARIAD SE

Inventor： Michael Holicki

IPC: G06V20/58 ,  G06T7/73 ,  B60W30/095

CPC classification number: G06V20/58 ,  G06T7/73 ,  B60W30/0956 ,  B60W2420/42

Abstract: A method for localizing a motor vehicle in an environment during a driving operation comprises, by a processor circuit in repeated estimation cycles, receiving sensor data of landmarks of the environment from an environment sensor, and ascertaining a respective estimated position of the motor vehicle from feature data, which are formed of map data of a map region of the environment and of the sensor data, using an estimation module. A movement path is estimated from the positions of multiple of the estimation cycles using a statistical observer model, and the observer model, during the formation of the movement path, models measurement noise contained in the position data as a covariance matrix of position coordinates of the position data, with matrix values of the covariance matrix being ascertained by the estimation module as a function of the sensor data and the map data.

公开(公告)号：US20230011156A1

公开(公告)日：2023-01-12

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 ,  G06V20/56 ,  G06V20/58 ,  G01S13/931

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：US20230288210A1

公开(公告)日：2023-09-14

申请号：US17998143

申请日：2021-05-04

Applicant: CARIAD SE

Inventor： Michael Holicki

IPC: G01C21/28

CPC classification number: G01C21/28

Abstract: A method is described in which a position of a vehicle is determined in a road network, having a plurality of edges, each of them being associated with a road section or a driving lane on a road section, by means of a sensor group of the vehicle, wherein first sensor data are detected by means of the sensor group, on the basis of the first sensor data each edge of a plurality of first edges of the road network is associated with a first probability that the vehicle is located in the road section associated with the first edge or in the driving lane associated with the first edge, second sensor data are detected by means of the sensor group, on the basis of the second sensor data and the probabilities associated with the first edges, a second probability is associated with each edge of a plurality of second edges of the road network that the vehicle is located in the road section associated with the second edge or in the driving lane associated with the second edge, and the second edge with the highest associated probability is determined to be the position of the vehicle.

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：US20230079899A1

公开(公告)日：2023-03-16

申请号：US17899361

申请日：2022-08-30

Applicant: CARIAD SE ,  Volkswagen Aktiengesellschaft

Inventor： Michael Holicki ,  Ralph Hänsel ,  Timo Iken ,  Roland Kube ,  Carolin Last ,  Stefan Wappler

IPC: G01C21/30

Abstract: To determine an absolute position of a vehicle, the vehicle is moved along a trajectory starting from the absolute initial position. The trajectory is sensed by odometry starting from a defined initial position, which is associated with the absolute initial position. Three objects are detected from the vehicle which is on the trajectory. At each detection, a detection beam is sensed. The trajectory with the detection beams are matched with a map. Finally, the absolute initial position is determined from the matching.

公开(公告)号：US20230350418A1

公开(公告)日：2023-11-02

申请号：US17786802

申请日：2020-12-04

Applicant: Cariad SE

Inventor： Michael Holicki

IPC: G05D1/02 ,  G01C21/00 ,  G06V10/774 ,  G06V20/56 ,  G06V10/82

CPC classification number: G05D1/0246 ,  G01C21/3848 ,  G06V10/774 ,  G06V20/56 ,  G06V10/82 ,  G06V2201/12

Abstract: A method for training an artificial neural generator network for the generating of synthetic landmark images is provided, in which landmark images are extracted from at least one training map as training data, which form a first training data set, and the generator network as a generator of a generative adversarial network learns with the aid of the first training data set and with the aid of real, non-annotated image data, recorded by a sensor device, to generate synthetic landmark images which are suited to reproducing the probability distribution underlying the first training data set. The invention also relates to a method for training an artificial neural localization network by the trained generator network and a method for determining a position of a mobile unit with at least one sensor device and at least one environment map by the trained localization network.

公开(公告)号：US12078491B2

公开(公告)日：2024-09-03

申请号：US17861035

申请日：2022-07-08

Applicant: Cariad SE ,  Volkswagen Aktiengesellschaft

Inventor： Michael Holicki ,  Ralph Hänsel ,  Timo Iken ,  Christian Jördens ,  Roland Kube ,  Carolin Last ,  Stefan Wappler

IPC: G01C21/30 ,  G06T7/73

CPC classification number: G01C21/30 ,  G06T7/73 ,  G06T2207/30256

Abstract: According to a method for self-localization of a vehicle, a first pose of the vehicle is determined. Environment sensor data are generated by means of an environment sensor device and on the basis of this a landmark is detected in the environment. A position of the landmark is determined and in dependence on this a second pose of the vehicle is determined. A memorized assignment instruction is consulted, matching up the first pose with at last one preferred sensor type or at least one dominant landmark type, and depending on this a first portion of the environment sensor data is selected that was generated by means of a first environment sensor system of the environment sensor device configured according to a first sensor type. For the detecting of the landmark, a first landmark detection algorithm is applied to the first portion of the environment sensor data.

公开(公告)号：US20230009978A1

公开(公告)日：2023-01-12

申请号：US17861035

申请日：2022-07-08

Applicant: Cariad SE ,  Volkswagen Aktiengesellschaft

Inventor： Michael Holicki ,  Ralph Hänsel ,  Timo Iken ,  Christian Jördens ,  Roland Kube ,  Carolin Last ,  Stefan Wappler

IPC: G01C21/30 ,  G06T7/73

Abstract: According to a method for self-localization of a vehicle, a first pose of the vehicle is determined. Environment sensor data are generated by means of an environment sensor device and on the basis of this a landmark is detected in the environment. A position of the landmark is determined and in dependence on this a second pose of the vehicle is determined. A memorized assignment instruction is consulted, matching up the first pose with at last one preferred sensor type or at least one dominant landmark type, and depending on this a first portion of the environment sensor data is selected that was generated by means of a first environment sensor system of the environment sensor device configured according to a first sensor type. For the detecting of the landmark, a first landmark detection algorithm is applied to the first portion of the environment sensor data.

公开(公告)号：US20230009073A1

公开(公告)日：2023-01-12

申请号：US17860930

申请日：2022-07-08

Applicant: Cariad SE

Inventor： Michael Holicki ,  Ralph Hänsel ,  Timo Iken ,  Roland Kube ,  Carolin Last ,  Stefan Wappler

IPC: G08G1/14 ,  G06V20/58

Abstract: According to a method for self-localization, a vehicle is detected by a vehicle-external sensor system of a parking infrastructure when it is located in a detection region of the vehicle-external sensor system. By way of a vehicle-external communication interface of the parking infrastructure, position information regarding the detection region is relayed in dependence on the detection of the vehicle to at least one vehicle computer unit and an initial pose of the vehicle is determined by the at least one vehicle computer unit in dependence on the position information.