Apparatus and method for recognizing travel lane

• US 10,635,911 B2
• Filed: 01/15/2018
• Issued: 04/28/2020
• Est. Priority Date: 01/16/2017
• Status: Active Grant

• Current Assignee: DENSO Corporation, Soken, Inc.
• Original Assignee: DENSO Corporation
• Inventors: Kumano, Shunya, Nitanda, Naoki, Watanabe, Akihiro
• Primary Examiner(s): Nguyen, Leon Viet Q
• Application Number:
  US15/871,769
• Publication Number:
  US 20180204075A1
• Time in Patent Office: 834 Days
• CPC Class Codes: B60W 2552/30 Road curve radius
• CPC Class Codes: B60W 30/12 Lane keeping
• CPC Class Codes: B60W 40/072 Curvature of the road
• CPC Class Codes: G06T 2207/30256 Lane; Road marking
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06V 20/588 Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
• CPC Class Codes: G08G 1/167 Driving aids for lane monitoring, lane changing, e.g. blind spot detection

Abstract

In an apparatus for recognizing a travel lane of a vehicle, a deviation determiner is configured to, if two or more shape change points are extracted by a shape change point extractor and the second derivative value of curvature of an extracted boundary line at at least one of the two or more shape change points is inverted in sign, determine whether or not the extracted boundary line and an estimated boundary line estimated from travel lane parameters estimated by a travel lane parameter estimator are deviating from each other beyond a predetermined allowable range. A driving aid is configured to, if it is determined that the extracted boundary line and the estimated boundary line are deviating from each other beyond the predetermined allowable range, perform control upon deviation to prevent occurrence of undesirable situations that may be caused by deviation between the extracted boundary line and the estimated boundary line.

Claims

1. An apparatus for recognizing a travel lane of a vehicle, comprising:

an image acquirer configured to acquire, from a camera mounted in the vehicle, an image of the travel lane ahead of the vehicle;

an edge extractor configured to extract boundary edge points used to estimate a boundary of the travel lane from the image acquired from the image acquirer;

a travel lane parameter estimator configured to use a filter prepared in advance to estimate travel lane parameters representing a condition and shape of the travel lane from coordinates of the boundary edge points extracted by the edge extractor;

a shape change point extractor configured to extract a shape change point at which an absolute value of a second derivative value of curvature of an extracted boundary line that is a boundary line represented by the boundary edge points extracted by the edge extractor exceeds a predetermined threshold;

a deviation determiner configured to, if two or more shape change points are extracted by the shape change point extractor and the second derivative value of curvature of the extracted boundary line at at least one of the two or more shape change points is inverted in sign, determine whether or not the extracted boundary line and an estimated boundary line that is a boundary line estimated from the travel lane parameters estimated by the travel lane parameter estimator are deviating from each other beyond a predetermined allowable range; and

a driving aid configured to, if it is determined by the deviation determiner that the extracted boundary line and the estimated boundary line are deviating from each other beyond the predetermined allowable range, perform specific control of the vehicle to prevent occurrence of undesirable situations caused by deviation between the extracted boundary line and the estimated boundary line.

2. The apparatus according to claim 1, wherein the driving aid is configured to, as the specific control, prohibit the estimated boundary line from being used further away from the vehicle than a sign-inverted shape change point that is, as viewed from the vehicle, a closest shape change point at which the second derivative value of curvature of the extracted boundary line is inverted in sign.

3. The apparatus according to claim 1, wherein the driving aid is configured to, as the specific control, cause the travel lane parameter estimator to estimate the travel parameters in each segment between the j-th closest sign-inverted shape change point and the (j+1)-th closest sign-inverted shape change point along the extracted boundary line, j being a positive integer value, the j-th closest sign-inverted shape change point being, as viewed from the vehicle, the j-th closest shape change point at which the second derivative value of curvature of the extracted boundary line is inverted in sign.

4. The apparatus according to claim 2, wherein:

the closest shape change point at which the second derivative value of curvature of the extracted boundary line is inverted in sign is a determination object point; and

the deviation determiner is configured to, if a distance between the determination object point along the extracted boundary line and the estimated boundary line in a lane width-wise direction exceeds a predetermined deviation threshold, determine that the extracted boundary line and the estimated boundary line are deviating from each other beyond the predetermined allowable range.

5. The apparatus according to claim 4, wherein the deviation threshold increases with increasing distance from the vehicle to the determination object point.

6. The apparatus according to claim 1, wherein, when both the extracted boundary line and the estimated boundary line are acquired for each of left and right boundaries of the travel lane, the deviation determiner is configured to, if a deviation direction of the left estimated boundary line from the left extracted boundary line and a deviation direction of the right estimated boundary line from the right extracted boundary line are the same direction, determine that the extracted boundary line and the estimated boundary line are deviating from each other beyond the predetermined allowable range.

7. The apparatus according to claim 1, wherein;

a straight line connecting a point closest to the vehicle along the extracted boundary line and a point furthest away from the vehicle along the same extracted boundary line is a reference line for the extracted boundary line; and

the shape change point extractor is configured to extract a maximal point along the extracted boundary line such that a distance from the maximal point to the reference line for the extracted boundary line becomes maximal, as the shape change point.

8. The apparatus according to claim 1, wherein the filter used by the travel lane parameter estimator is a Kalman filter.

17. The apparatus according to claim 1, wherein the specific control includes at least notification control to notify a driver of the vehicle that estimation accuracy of the travel lane parameters has been decreased or functionality suspension control to suspend some or all of control functions using the travel lane parameters.

9. A method of recognizing a travel lane of a vehicle, comprising:

acquiring from a camera mounted in the vehicle an image of the travel lane;

extracting boundary edge points of the travel lane;

using a filter prepared in advance to estimate travel lane parameters representing a condition and shape of the travel lane from coordinates of the boundary edge points;

extracting a shape change point at which an absolute value of a second derivative value of curvature of an extracted boundary line represented by the boundary edge points extracted exceeds a predetermined threshold;

acquiring an estimated boundary line that is a boundary line estimated from the travel lane parameters;

determining whether or not the extracted boundary line and the estimated boundary line are deviating from each other beyond a predetermined allowable range, if two or more shape change points are extracted by the shape change point extractor and the second derivative value of curvature of the extracted boundary line at at least one of the two or more shape change points is inverted in sign; and

performing specific control of the vehicle to prevent occurrence of undesirable situations caused by deviation between the extracted boundary line and the estimated boundary line, if it is determined that the extracted boundary line and the estimated boundary line are deviating from each other beyond the predetermined allowable range.

10. The method according to claim 9, wherein performing the specific control includes, as the specific control, prohibiting the estimated boundary line from being used further away from the vehicle than a sign-inverted shape change point that is, as viewed from the vehicle, a closest shape change point at which the second derivative value of curvature of the extracted boundary line is inverted in sign.

11. The method according to claim 9, wherein performing the specific control includes, as the specific control, estimating the travel parameters in each segment between the j-th closest sign-inverted shape change point and the (j+1)-th closest sign-inverted shape change point along the extracted boundary line, j being a positive integer value, the j-th closest sign-inverted shape change point being, as viewed from the vehicle, the j-th closest shape change point at which the second derivative value of curvature of the extracted boundary line is inverted in sign.

12. The method according to claim 10, wherein:

the closest shape change point at which the second derivative value of curvature of the extracted boundary line is inverted in sign is a determination object point; and

determining whether or not the extracted boundary line and the estimated boundary line are deviating from each other includes, if a distance between the determination object point along the extracted boundary line and the estimated boundary line in a lane width-wise direction exceeds a predetermined deviation threshold, determining that the extracted boundary line and the estimated boundary line are deviating from each other beyond the predetermined allowable range.

13. The method according to claim 12, wherein the deviation threshold increases with increasing distance from the vehicle to the determination object point.

14. The method according to claim 9, wherein, when both the extracted boundary line and the estimated boundary line are acquired for each of left and right boundaries of the travel lane, determining whether or not the extracted boundary line and the estimated boundary line are deviating from each other includes, if a deviation direction of the left estimated boundary line from the left extracted boundary line and a deviation direction of the right estimated boundary line from the right extracted boundary line are the same direction, determining that the extracted boundary line and the estimated boundary line are deviating from each other beyond the predetermined allowable range.

15. The method according to claim 9, wherein:

a straight line connecting a point closest to the vehicle along the extracted boundary line and a point furthest away from the vehicle along the same extracted boundary line is a reference line for the extracted boundary line; and

extracting a shape change point includes extracting a maximal point along the extracted boundary line such that a distance from the maximal point to the reference line for the extracted boundary line becomes maximal, as the shape change point.

16. The method according to claim 9, wherein the filter is a Kalman filter.

18. The method according to claim 9, wherein the specific control includes at least notification control to notify a driver of the vehicle that estimation accuracy of the travel lane parameters has been decreased or functionality suspension control to suspend some or all of control functions using the travel lane parameters.