Method for the anticipated ascertainment of a bend on a portion of road, and associated system

• US 8,103,422 B2
• Filed: 09/25/2007
• Issued: 01/24/2012
• Est. Priority Date: 09/26/2006
• Status: Active Grant

• Current Assignee: Valeo Vision (Valeo SE)
• Original Assignee: Valeo Vision (Valeo SE)
• Inventors: Herbin, Anne, Basset, Michel, Lauffenburger, Jean-Philippe, Bradai, Benazouz
• Primary Examiner(s): Trammell, James
• Assistant Examiner(s): Shafi, Muhammad
• Application Number:
  US11/860,681
• Publication Number:
  US 20080249706A1
• Time in Patent Office: 1,582 Days
• Field of Search: 701/58, 701/77, 701/36, 701/49, 701200-226, 340/437, 340/458, 340/565, 340/468, 340/555, 340/556, 362/466, 362/465, 382/316, 382/321
• US Class Current: 701/77
• CPC Class Codes: B60Q 1/085 due to special conditions, e.g. adverse weather, type of road, badly illuminated road signs or potential dangers B60Q1/10, B60Q1/12, B60Q1/1423 take precedence
• CPC Class Codes: B60Q 1/12 due to steering position
• CPC Class Codes: B60Q 2300/112 Vehicle speed
• CPC Class Codes: B60Q 2300/322 Road curvature
• CPC Class Codes: B60Q 2300/33 Driving situation
• CPC Class Codes: B60W 2554/00 Input parameters relating to objects
• CPC Class Codes: B60W 2556/50 of positioning data, e.g. GPS [Global Positioning System] data
• CPC Class Codes: B60W 40/072 Curvature of the road
• CPC Class Codes: G01C 21/26 specially adapted for navigation in a road network
• CPC Class Codes: G01C 21/3697 Output of additional, non-guidance related information, e.g. low fuel level G01C21/3679 takes precedence

Abstract

A system and method for the anticipated detection of a bend on a portion of road taken by a motor vehicle comprising the following steps: establishing, by means of a first system, called as the navigation system, a first set of information on the bend in question, the first set of information being associated with a first confidence index; establishing, by means of a second system, called as the image processing system, a second set of information on the bend in question, the second set of information being associated with a second confidence index; and establishing, from the first set of information and the second set of information and by taking into account the first confidence index and the second confidence index, a third set of information on the bend in question.

Claims

1. A method for the anticipated detection of a bend on a portion of road taken by a motor vehicle, comprises the following steps:

establishing a first set of information on the bend in the portion of the road in question utilizing a navigation system, said first set of information being associated with a first confidence index, said navigation system comprising a receiver for receiving data from a geographical positioning system and mapping data;

establishing a second set of information on the bend in question utilizing an image processing system, said second set of information being associated with a second confidence index, said image processing system comprising a camera and an image processing application that captures an image on the bend; and

establishing a third set of information on the bend in question using a data fusion system questioning the portion of the road in response to the first set of information and the second set of information, each of said first and second sets of information comprising data on at least one of a radius of curvature of the bend in the portion of the road or the distance between the bend and the vehicle;

using said third set of information to determine before the bend has been entered, a relevant light point in the bend, and the movement of said relevant light point between entering the bend and leaving the bend and moving the headlights through an angle in response thereto.

2. The method according to claim 1, wherein said third set of information is associated with a third confidence index developed from said first confidence index and said second confidence index.

3. The method according to claim 2, wherein said first confidence index is developed from at least two parameters of a first set of parameters comprising the following parameters:

accuracy of the GPS positioning;

accuracy of the digitization of the mapping;

date on which the mapping was updated;

type of road;

level of information on the road;

functional class of the road;

environment of the vehicle;

selection of a guide mode by the driver and level of conformity between a planned route and information provided by sensors on board the vehicle.

4. The method according to claim 3, wherein said first confidence index is developed by taking a weighted average of values allocated to the following parameters, said parameters being associated with weighting coefficients resulting from a learning phase.

5. The method according to claim 2, wherein said second confidence index is developed from one or more parameters, of a second set of parameters relating to an image obtained by the camera, from among at least two of the following parameters:

measurement of the texture of the image in question;

shadow factor on the image in question;

vertical gradient of the decrease in light;

index of symmetry of the image in question.

6. The method according to claim 5, wherein said second confidence index is developed by taking a weighted average of values allocated to all of the parameters of said second set of parameters, said parameters being associated with weighting coefficients resulting from a learning phase.

7. The method according to claim 2, wherein it includes the additional steps comprising:

comparing the first confidence index to a first threshold value and the second confidence index to a second threshold value;

considering, when establishing the third set of information, only the set(s) of information, among the first set of information and the second set of information, of which the confidence index is greater than the threshold value to which it is compared.

8. The method according to claim 7, wherein it includes the additional step of adopting, in the event of said first confidence index and said second confidence index being less than the threshold values to which they are compared, a fail-safe mode of operation in which the bend in question is detected instantaneously.

9. The method according to claim 1, wherein said first set of information and said second set of information comprise at least one item of data from among the following:

radius of curvature of the bend in question, distance between a momentary position of the vehicle and the entering of the bend in question, type of bend.

10. The method according to claim 1, wherein said third set of information comprises at least one item of data from among the following:

radius of curvature of the bend in question, distance between a momentary position of the vehicle and the entering of the bend in question, type of bend, relevant lighting point.

11. The method according to claim 1, wherein said third set of information comprises data, the value of each of said items of data being established by taking a weighted average of corresponding data from said first set of information and said second set of information, an item of data of the first set being weighted by a first weighting coefficient and the corresponding data of said second set of information being weighted by a second weighting coefficient, said first weighting coefficient being greater than said second weighting coefficient if and only if said first confidence index is greater than said second confidence index.

12. The method according to claim 1, wherein said first set of information and said second set of information are processed jointly in a system for processing information by data fusion.

13. The method according to claim 1, wherein the image processing system carries out the operation consisting in detecting curbs of the portion of road in question.

14. The method according to claim 1, wherein said receiver for data from a geographical positioning system is a GPS aerial.