Method for lane detection

• US 10,115,026 B2
• Filed: 03/27/2015
• Issued: 10/30/2018
• Est. Priority Date: 04/04/2014
• Status: Active Grant

• Current Assignee: Aptiv Technologies AG
• Original Assignee: Delphi Technologies, Inc. (BorgWarner Incorporated)
• Inventors: Mueter, Mirko, Zhao, Kun, Szumilas, Lech J., Mueller, Dennis, Nunn, Christian
• Primary Examiner(s): Tran, Thai
• Assistant Examiner(s): MESA, JOSE M
• Application Number:
  US15/303,385
• Publication Number:
  US 20170068862A1
• Time in Patent Office: 1,313 Days
• Field of Search: 348148
• CPC Class Codes: G06T 7/277 involving stochastic approaches, e.g. using Kalman filters
• 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: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

A method for lane detection for a camera-based driver assistance system includes the following steps: image regions in images that are recorded by a camera are identified as detected lane markings if the image regions meet a specified detection criterion. At least two detected lane markings are subjected to a tracking process as lane markings to be tracked. By means of a recursive state estimator, separate progressions are estimated for at least two of the lane markings to be tracked. Furthermore, for each of a plurality of the detected lane markings, a particular offset value is determined, which indicates a transverse offset of the detected lane marking in relation to a reference axis. By means of an additional estimation method, the determined offset values are each associated with one of the separate progressions of the lane markings to be tracked.

Claims

1. A method for lane detection for a camera-based driver assistance system comprising a camera attached to a vehicle for capturing images of a traffic space ahead of the vehicle, wherein the method includes:

a) identifying in an image captured by a camera that areas of the image are identified as recognized lane markings which meet a predetermined recognition criterion;

b) tracking at least two recognized lane markings, wherein a temporal change of a course of the at least two lane markings to be tracked in the road surface plane is tracked using a common recursive state estimator,

c) estimating separate courses for the at least two of the lane markings to be tracked;

d) determining a respective offset value for the at least two recognized lane markings, wherein said respective offset value indicates a transversal offset of the at least two recognized lane markings relative to a reference axis parallel to a center longitudinal axis of the vehicle; and

e) performing an allocation of the respective offset value to a respective one of the separate courses of the lane markings to be tracked, wherein for allocation of the determined offset values to a respective one of the separate courses of the lane markings to be tracked, a parametric estimation method is applied; and

f) outputting a lane-departure-warning in accordance with the offset values.

2. The method according to claim 1, wherein separate courses are estimated for exactly two of the lane markings to be tracked, wherein the allocation of the determined offset values to a respective one of the two separate courses is indicated by a set of binary parameters.

3. The method according to claim 2, wherein the set of binary parameters can be used for parameterization of a model curve approximately describing the course of the lane.

4. The method according to claim 2, wherein given combinations of values of the set of binary parameters are used as input variables for generating additional estimation hypotheses.

5. The method according to claim 4, wherein the tracking process is performed separately for several different hypotheses and following a predetermined number of computing cycles, a most likely hypothesis is determined for which a deviation between the estimated courses of lane markings to be tracked and the courses of recognized lane markings is least.

6. The method according to claim 5, wherein only two tracking processes for different hypotheses are performed simultaneously in single comparison steps, respectively.

7. The method according to claim 6, wherein starting with an arbitrary start hypothesis several single comparison steps are carried out successively, wherein the most likely hypothesis of a previous single comparison step is respectively compared with a hypothesis not yet subjected to a single comparison step.

8. The method according to one of claim 4, wherein for every generated hypothesis a separate allocation of measurement values to the lane markings is performed using the Hungarian method.

9. The method according to claim 1, wherein respective offset values are determined for at least four recognized lane markings.

10. The method according to claim 1, wherein a test is made before performing the additional estimation method when for each separately estimated course at least one or two offset values are determinable, wherein otherwise the state estimator is limited by means of a logic based on the number of offset values.

11. The method according to claim 1, wherein the state estimator uses a state vector whose components comprise course parameters for each of the lane markings to be tracked and additionally several offset parameters formed by the offset values, wherein, preferably, the state estimator uses a state vector whose components additionally include an angular offset of the vehicle relative to one of the two separate courses, an angular offset of the vehicle relative to the other of the two separate courses, an angle of inclination of the road surface plane relative to an average road surface plane and/or a lateral speed of the vehicle.

12. The method according to claim 1, wherein the reference axis coincides with the center longitudinal axis of the vehicle or with its vertical projection on the road surface plane and the offset values each indicate a transverse offset of a lane marking in the road surface plane.

13. The method according to claim 1, wherein only those two lane markings are subjected to the tracking process which correspond to a left and a right lane boundary of the own-lane driven on by the vehicle.