Method and apparatus for guiding a vehicle in the surroundings of an object

Patent metadata

Field	Value
Publication number	US-9734719-B2
Application number	US-201414781986-A
Country	US
Kind code	B2
Filing date	Feb 20, 2014
Priority date	Apr 3, 2013
Publication date	Aug 15, 2017
Grant date	Aug 15, 2017

Abstract

A method for guiding a vehicle in the surrounding environment of an object. The method includes reading in a multiplicity of view ray endpoints in a three-dimensional image, produced by a stereo camera of the vehicle, of a surrounding environment, containing the object, of the vehicle, at least one of the view ray endpoints representing an outer surface of the object, connecting the multiplicity of view ray endpoints to form a polygon that represents a free surface to be traveled by the vehicle, and generating a driving corridor, provided for the vehicle, for driving around the object, based on the free surface.

First claim

What is claimed is:

1. A method for guiding a vehicle in a surrounding environment of an object, the method comprising:
reading in a multiplicity of view ray endpoints in a three-dimensional image of the surrounding environment, containing the object and/or the vehicle, the three-dimensional image representing an image produced by a stereo camera of the vehicle, and at least one of the view ray endpoints representing an outer surface of the object;
connecting the multiplicity of view ray endpoints to form a polygon that represents a free surface to be traveled by the vehicle, wherein one or more sides of the polygon define an edge of the free space, the one or more sides defining a boundary between the polygon and the object; and
generating a driving corridor, provided for the vehicle, for driving around the object, based on the free surface, wherein the generating of the driving corridor includes:
determining a safety distance between the vehicle and the one or more sides of the polygon, wherein a size of the safety distance varies as a function of a classification of the object, and
defining at least one lateral edge of the driving corridor, a distance between the at least one lateral edge of the driving corridor and the one or more sides of the polygon corresponding to the safety distance.

2. The method of claim 1 , further comprising:
producing the three-dimensional image in the form of a disparity map based on a first image and a second image of the stereo camera, a specified distance value being assigned to each of a multiplicity of points of the three-dimensional image.

3. The method of claim 2 , further comprising:
determining a selection of points from the multiplicity of points of the three-dimensional image, which represent an end of a region of view of the stereo camera, as the multiplicity of view ray endpoints of the three-dimensional image.

4. The method of claim 3 , wherein in the determining, the selection of the points takes place through line-by-line scanning and column-by-column combining of the multiplicity of points of the three-dimensional image which represent the end of the region of view.

5. The method of claim 4 , wherein in the determining, in the column-by-column combining a relative speed between the vehicle and the end of the region of view is taken into account to compensate a line-by-line offset of the points inside the three-dimensional image.

6. The method of claim 1 , wherein a time-based filtering is used in the connecting or in the determining.

7. The method of claim 6 , wherein a Kalman filter performs the time-based filtering.

8. The method of claim 1 , wherein the generated driving corridor has left and right lateral edges respectively corresponding to left and right sides of the vehicle.

9. The method of claim 1 , wherein a first lateral edge of the driving corridor is generated based on obstacles on a first side of the vehicle, and a second lateral edge of the driving corridor is generated based on obstacles on a second side of the vehicle, opposite to the first side.

10. The method of claim 1 , further comprising:
performing line-by-line scanning of the three-dimensional image and column-by-column combining of a multiplicity of points of the three-dimensional image;
determining a selection of the multiplicity of view ray endpoints of the three-dimensional image which represent the object based on the line-by-line scanning of the three-dimensional image and the column-by-column combining of the multiplicity of points of the three-dimensional image; and
compensating a line-by-line offset of the points inside the three-dimensional image by taking into account a relative speed between the vehicle and the object in the column-by-column combining of the multiplicity of points.

11. The method of claim 1 , wherein a specified color value is assigned to each specified distance value to represent a spatial depth of the three-dimensional image.

12. A device for guiding a vehicle in a surrounding environment of an object, comprising:
a read-in device configured to read in a multiplicity of view ray endpoints in a three-dimensional image of the surrounding environment, containing the object and/or the vehicle, the three-dimensional image representing an image produced by a stereo camera of the vehicle, and at least one of the view ray endpoints representing an outer surface of the object;
a connecting device configured to connect the multiplicity of view ray endpoints to form a polygon that represents a free surface to be traveled by the vehicle, wherein one or more sides of the polygon define an edge of the free space, the one or more sides defining a boundary between the polygon and the object; and
a generating device configured to generate a driving corridor, provided for the vehicle, for driving around the object, based on the free surface, wherein the generating of the driving corridor includes:
determining a safety distance between the vehicle and the one or more sides of the polygon, wherein a size of the safety distance varies as a function of a classification of the object, and
defining at least one lateral edge of the driving corridor, a distance between the at least one lateral edge of the driving corridor and the one or more sides of the polygon corresponding to the safety distance.

13. The device of claim 12 , wherein the generated driving corridor has left and right lateral edges respectively corresponding to left and right sides of the vehicle.

14. The method of claim 12 , wherein a first lateral edge of the driving corridor is generated based on obstacles on a first side of the vehicle, and a second lateral edge of the driving corridor is generated based on obstacles on a second side of the vehicle, opposite to the first side.

15. The device of claim 12 , further configured to:
perform line-by-line scanning of the three-dimensional image and column-by-column combining of a multiplicity of points of the three-dimensional image;
determine a selection of the multiplicity of view ray endpoints of the three-dimensional image which represent the object based on the line-by-line scanning of the three-dimensional image and the column-by-column combining of the multiplicity of points of the three-dimensional image; and
compensate a line-by-line offset of the points inside the three-dimensional image by taking into account a relative speed between the vehicle and the object in the column-by-column combining of the multiplicity of points.

16. A non-transitory computer readable medium having a computer program, which is executable by a processor, comprising:
a program code arrangement having program code for computer program product having program code guiding a vehicle in a surrounding environment of an object, by performing the following:
reading in a multiplicity of view ray endpoints in a three-dimensional image of the surrounding environment, containing the object and/or the vehicle, the three-dimensional image representing an image produced by a stereo camera of the vehicle, and at least one of the view ray endpoints representing an outer surface of the object;
connecting the multiplicity of view ray endpoints to form a polygon that represents a free surface to be traveled by the vehicle, wherein one or more sides of the polygon define an edge of the free space, the one or more sides defining a boundary between the polygon and the object; and
generating a driving corridor, provided for the vehicle, for driving around the object, based on the free surface, wherein the generating of the driving corridor includes:
determining a safety dist

Assignees

• Bosch Gmbh Robert

Inventors

• Flehmig Folko
• Gussner Thomas
• Ma Yuefeng

Classifications

• G08G1/167 Primary

  Driving aids for lane monitoring, lane changing, e.g. blind spot detection · CPC title

• H04N13/204

  using stereoscopic image cameras (stereoscopic photography G03B35/00) · CPC title

• G08G1/165

  for passive traffic, e.g. including static obstacles, trees · CPC title

• G08G1/166

  for active traffic, e.g. moving vehicles, pedestrians, bikes · CPC title

• H04N13/0203

  Electricity · mapped topic