METHOD OF DETECTING OBJECT IN THE VICINITY OF A VEHICLE

• US 20110098892A1
• Filed: 06/25/2008
• Published: 04/28/2011
• Est. Priority Date: 06/25/2008
• Status: Active Grant

• Current Assignee: Arriver Software AB (Qualcomm, Inc.)
• Original Assignee: Autoliv AB (Autoliv Incorporated)
• Inventors: Hanqvist, Mattias, Lundmark, Astrid
• Granted Patent: US 8,525,728 B2
• Time in Patent Office: Days
• US Class Current: 701/45
• CPC Class Codes: B60R 2021/01272 Belt tensioners
• CPC Class Codes: B60R 21/0134 responsive to imminent contact with an obstacle , e.g. using radar systems
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads

Abstract

A system for detecting objects in the vicinity of a vehicle, comprising: a sensor for gathering data relating to objects in the vicinity of the vehicle; and a processor operable to: detect objects, from the sensor data, based on one or more detection criteria, the detection being regulated by one or more detection parameters; analyse the detected objects; and in response to the number of detected objects, vary one or more of the detection parameters.

Claims

1. A system for detecting at least one object in the vicinity of a vehicle, comprising:

a sensor for gathering data relating to the object in the vicinity of the vehicle; and

a processor operable to;

detect object from the data from the sensor, based on one or more detection criteria, the detection criteria being regulated by one or more detection parameters;

analyse the detected object; and

in response to the number of the object detected, vary one or more of the detection parameters.

2. A system according to claim 1, wherein the sensor comprises a video camera.

3. A system according to claim 1, wherein at least one of the detection parameters comprises a threshold relating to the angular size of the object appearing in the data gathered from the sensor.

4. A system according to claim 1, wherein at least one of the detection parameters relates to the distance of the object from the vehicle.

5. A system according to claim 1, wherein at least one of the detection parameter relates to the contrast between the object and its background.

6. A system according to claim 1, wherein at least one of the detection parameter relates to the position of the object with respect to the vehicle.

7. A system according to claim 6, wherein the at least one detection parameter relates to the angular distance of a line joining the vehicle and the object from the trajectory of the vehicle.

8. A system according to claim 1, wherein the at least one detection parameter relates to the type of the object that is detected.

9. A system according to claim 1, wherein the at least one detection parameter relates to an estimated probability that the object is of a specific type.

10. A system according to claim 1, wherein the at least one detection parameter relates to the application of a filter to an image or other of the data gathered by the sensor.

11. A system according to claim 10, wherein the application of the filter smoothes out small features in the image or other of the data.

12. A system according to claim 1, wherein:

a preferred number of detections of the object is defined for a set of the data gathered by the sensor; and

the at least one detection parameter is varied so that the number of the detections produced using the detection parameters for a subsequent set of data is close to the preferred number.

13. A system according to claim 1, wherein the derivative of the number of the detections with respect to variation of at least one of the detection parameters is determined.

14. A system according to claim 1, wherein the detected object is analysed by a classifier which is trained through analysis of a large number of known types of the objects.

15. A system according to claim 1, wherein the detected object is analysed to reach a determination as to whether the vehicle is likely to be involved in an impact with the object.

16. A system according to claim 15 wherein, if it appears that the vehicle is likely to be involved in an impact with one of the objects, a vehicle safety system is activated.

17. A system according to claim 1, further comprising a display on which at least some of the data gathered by the sensor is displayed to a vehicle occupant.

18. A system according to claim 17, wherein the object that is determined to be potentially relevant is highlighted in the display.

19. A system according to any claim 1, wherein the detected object is classified by a classifier having two or more classification steps.

20. A system according to claim 19, wherein a first classification of the two or more steps is regulated by one or more classification parameters, and the one or more of the classification parameters is varied in response to the number of the object classified in the first classification step.