Method and apparatus for segmenting an occupancy grid for a surroundings model of a driver assistance system for a vehicle background of the invention

• US 10,089,419 B2
• Filed: 11/21/2014
• Issued: 10/02/2018
• Est. Priority Date: 11/21/2013
• Status: Active Grant

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Michalke, Thomas, Glaeser, Claudius, Buerkle, Lutz
• Primary Examiner(s): Mudrick, Timothy A
• Application Number:
  US14/550,093
• Publication Number:
  US 20150154328A1
• Time in Patent Office: 1,411 Days
• Field of Search: None
• CPC Class Codes: B60W 40/04 Traffic conditions
• CPC Class Codes: G06F 30/15 Vehicle, aircraft or watercraft design
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30

Abstract

A method is described for segmenting an occupancy grid for a surroundings model of a driver assistance system for a vehicle. The method encompasses a step of reading in an occupancy grid having a plurality of grid cells, each grid cell of the plurality of grid cells having assigned to it a grid cell information item that encompasses an information item about a degree of occupancy and at least one additional information item about the grid cell; and a step of assigning at least one object and/or a plurality of objects to the plurality of grid cells using the grid cell information item, in order to segment the occupancy grid for a surroundings model.

Claims

1. A method for segmenting an occupancy grid for a surroundings model of a driver assistance system for a vehicle, comprising:

reading in an occupancy grid having a plurality of grid cells, each grid cell of the plurality of grid cells having assigned thereto a grid cell information item that encompasses an information item about a degree of occupancy and at least one additional information item about the grid cell;

assigning at least one of at least one object and a plurality of objects to the plurality of grid cells using the grid cell information item, in order to segment the occupancy grid for a surroundings model; and

predicting object assignments for the plurality of grid cells after the reading-in step, the grid cell information item for each grid cell encompassing a predicted object assignment, wherein in the predicting step the object assignment is predicted using at least one of an object assignment and a grid cell information item from a previous execution of the method.

2. The method as recited in claim 1, further comprising:

projecting the plurality of grid cells as transformed grid cells into a feature space using the grid cell information item, wherein in the assigning step the at least one of the at least one object and the plurality of objects is assigned to the plurality of grid cells using one of the transformed grid cells in the feature space and an information item derived therefrom.

3. The method as recited in claim 1, further comprising:

extracting proximity relationships between the transformed grid cells in a feature space and preparing a graph using the proximity relationships, nodes of the graph representing the grid cells, and edges between the nodes representing proximity relationships, wherein in the assigning step the at least one of the at least one object and the plurality of objects is assigned to the plurality of grid cells using at least one of the proximity relationships, the graph, and an information item derived therefrom.

4. The method as recited in claim 3, further comprising:

allocating a weight to each edge of the graph using the proximity relationships, the weight representing an affiliation of a node with an object.

7. The method as recited in claim 1, wherein, in the predicting step, for each grid cell of the plurality of grid cells, a grid cell information item is predicted using at least one of an object assignment and a grid cell information item from a previous execution of the method.

5. A method for segmenting an occupancy grid for a surroundings model of a driver assistance system for a vehicle, comprising:

reading in an occupancy grid having a plurality of grid cells, each grid cell of the plurality of grid cells having assigned thereto a grid cell information item that encompasses an information item about a degree of occupancy and at least one additional information item about the grid cell;

assigning at least one of at least one object and a plurality of objects to the plurality of grid cells using the grid cell information item, in order to segment the occupancy grid for a surroundings model;

extracting proximity relationships between the transformed grid cells in a feature space and preparing a graph using the proximity relationships, nodes of the graph representing the grid cells, and edges between the nodes representing proximity relationships, wherein in the assigning step the at least one of the at least one object and the plurality of objects is assigned to the plurality of grid cells using at least one of the proximity relationships, the graph, and an information item derived therefrom;

allocating a weight to each edge of the graph using the proximity relationships, the weight representing an affiliation of a node with an object; and

segmenting the graph using the weight of each edge nodes being combined into sub-graphs, a sub-graph encompassing only nodes of one object, wherein in the assigning step the at least one of the at least one object and the plurality of objects is assigned to the plurality of grid cells using at least one of the segmented graph, cliques, and an information item derived therefrom.

6. The method as recited in claim 5, further comprising:

after the segmenting step, performing a transforming step in which the grid cells of the occupancy grid are assigned to an object using the sub-graphs of the graph.

8. An apparatus for segmenting an occupancy grid for a surroundings model of a driver assistance system for a vehicle, the apparatus comprising:

an interface for reading in an occupancy grid having a plurality of grid cells, each grid cell of the plurality of grid cells having assigned thereto a grid cell information item that encompasses an information item about a degree of occupancy and at least one additional information item about the grid cell;

a device for assigning at least one of at least one object and a plurality of objects to the plurality of grid cells using the grid cell information item, in order to segment the occupancy grid for a surroundings model; and

a device for predicting object assignments for the plurality of grid cells after reading in the occupancy grid, the grid cell information item for each grid cell encompassing a predicted object assignment, wherein the device for predicting predicts the object assignment using at least one of an object assignment and a grid cell information item from a previous execution of the segmenting by the apparatus.

9. A non-transitory computer program product having program code for carrying out a method when the program product is executed on at least one of an apparatus, a computer, and a control unit, the method being for segmenting an occupancy grid for a surroundings model of a driver assistance system for a vehicle, the method comprising:

reading in an occupancy grid having a plurality of grid cells, each grid cell of the plurality of grid cells having assigned thereto a grid cell information item that encompasses an information item about a degree of occupancy and at least one additional information item about the grid cell;

assigning at least one of at least one object and a plurality of objects to the plurality of grid cells using the grid cell information item, in order to segment the occupancy grid for a surroundings model; and

predicting object assignments for the plurality of grid cells after the reading-in step, the grid cell information item for each grid cell encompassing a predicted object assignment, wherein in the predicting step the object assignment is predicted using at least one of an object assignment and a grid cell information item from a previous execution of the method.