Method and System for Determining Landmarks in an Environment of a Vehicle

Patent metadata

Field	Value
Publication number	US-2020200545-A1
Application number	US-201916724468-A
Country	US
Kind code	A1
Filing date	Dec 23, 2019
Priority date	Dec 21, 2018
Publication date	Jun 25, 2020
Grant date	—

Abstract

In the method for determining landmarks in an environment of a vehicle, initial environmental data are captured, and a least a subset of the initial environmental data is determined. The determined subset is assigned a category from a plurality of different categories. Depending on a landmark type, a category is determined, and landmark detection for the landmark type is carried out restricted to the subset of the determined category. The system for determining landmarks in an environment of a vehicle comprises a capturing unit for capturing initial environmental data, a segmenting unit for determining at least a subset of the initial environmental data and for assigning a category from a plurality of different categories to the determined subset, and a computing unit for determining a category depending on a landmark type, and for carrying out landmark detection for the landmark type. In so doing, the computing unit is configured to carry out landmark detection restricted to the subset of the determined category.

First claim

What is claimed is:

1 . A method for determining landmarks in an environment of a vehicle, wherein
initial environmental data are captured; at least a subset of the initial environmental data are determined; the determined subset is assigned a category from a plurality of different categories; depending on a landmark type, a category is determined; and landmark detection is carried out for the landmark type restricted to the subset of the determined category.

2 . The method according to claim 1 , wherein the initial environmental data are captured by a camera.

3 . The method of claim 1 , wherein the subset is determined using a machine learning method.

4 . The method of claim 1 , wherein the categories comprise a ground, a road, a vegetation, a developed region, a pillar, a stationary or driving vehicle, or a person.

5 . The method of claim 1 , wherein landmark detection is carried out using an heuristic method.

6 . The method of claim 1 , wherein
a spatial region in the environment of the vehicle is determined using the determined subset; and other environmental data are captured; wherein
to capture the other environmental data, another sensor type is used than for capturing the initial environmental data; and
the landmark detection is furthermore carried out using the other environmental data and depending on the spatial region.

7 . The method of claim 1 , wherein
the subset is transmitted by a mobile unit to an external unit; and landmark detection is carried out by the external unit.

8 . The method of claim 7 , wherein
request data are generated by the external unit; and the subset is transmitted to the external unit depending on the request data.

9 . The method of claim 1 , wherein
position data for landmarks of the landmark type are determined and output during landmark detection; wherein using the position data, a posture of the vehicle is determined.

10 . The method of claim 1 , wherein landmark detection is carried out using a machine learning method.

11 . The method of claim 10 , wherein
training data are generated using the subset; and by using the training data, landmark detection training is carried out using the machine learning method.

12 . A system for determining landmarks in an environment of a vehicle, comprising
a capturing unit for capturing initial environmental data; a segmenting unit for determining at least a subset of the initial environmental data and for assigning a category from a plurality of different categories to the determined subset; and a computing unit for determining a category depending on a landmark type, and for carrying out landmark detection for the landmark type; wherein the computing unit is configured to carry out landmark detection restricted to the subset of the determined category.

13 . The system of claim 12 , wherein the capturing unit comprises a camera.

14 . The system of claim 12 , wherein a controller actuates the segmenting unit so that only the subset of the captured environmental data is transmitted to the computing unit.

15 . The system of claim 12 , further comprising
an interface with an external unit; wherein the system is configured to transmit the determined subset to the external unit; wherein the external unit is configured to carry out landmark detection using the subset.

16 . The method of claim 2 , wherein the subset is determined using a machine learning method.

17 . The method of claim 2 , wherein the categories comprise a ground, a road, a vegetation, a developed region, a pillar, a stationary or driving vehicle, or a person.

18 . The method of claim 3 , wherein the categories comprise a ground, a road, a vegetation, a developed region, a pillar, a stationary or driving vehicle, or a person.

19 . The system of claim 13 , wherein a controller actuates the segmenting unit so that only the subset of the captured environmental data is transmitted to the computing unit.

20 . The system of claim 13 , further comprising
an interface with an external unit; wherein
the system is configured to transmit the determined subset to the external unit; wherein
the external unit is configured to carry out landmark detection using the subset.

Assignees

• Volkswagen Ag

Inventors

• Wappler Stefan
• Hüger Philipp

Classifications

• G06V20/588

  Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road · CPC title

• G06V20/582

  of traffic signs · CPC title

• G06V20/56 Primary

  exterior to a vehicle by using sensors mounted on the vehicle · CPC title

• G06V10/82

  using neural networks · CPC title

• G01C21/30 Primary

  Map- or contour-matching · CPC title