System, method, and computer-readable recording medium for lane keeping control

• US 9,469,343 B2
• Filed: 01/30/2014
• Issued: 10/18/2016
• Est. Priority Date: 02/07/2013
• Status: Active Grant

• Current Assignee: HL Klemove Corporation
• Original Assignee: Mando Corporation
• Inventors: Kim, Soon Tae
• Primary Examiner(s): Elahi, Shan
• Application Number:
  US14/168,577
• Publication Number:
  US 20140218508A1
• Time in Patent Office: 992 Days
• Field of Search: 348/118
• US Class Current: 1/1
• CPC Class Codes: B62D 15/025 Active steering aids, e.g. helping the driver by actively influencing the steering system after environment evaluation B62D1/28 takes precedence
• CPC Class Codes: G05D 1/0246 using a video camera in combination with image processing means
• 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

Abstract

A lane keeping control system includes: a lane parameter generation unit configured to generate a relative position between a lane and a vehicle as a lane parameter based on information received from a camera installed on the vehicle; a vehicle state information reception unit configured to receive vehicle state information sensed or measured by each vehicle sensor or device installed on the vehicle; an equivalent model generation unit configured to receive the lane parameter from the lane parameter generation unit, receive the vehicle state information from the vehicle state information reception unit, analyze a time flow-related correlation between the vehicle parameter and the vehicle state information, and generate an equivalent model of each; an erroneous detection determination unit configured to determine and classify a situation of erroneous detection of a lane parameter by simultaneously analyzing a case.

Claims

1. A lane keeping control system, comprising:

a fault detection device including;

a lane parameter generation unit that generates a relative position between a lane and a vehicle as a lane parameter, based on information received from a camera installed on the vehicle;

a vehicle state information reception unit that receives vehicle state information sensed or measured by each vehicle sensor or device installed on the vehicle;

an equivalent model generation unit that

receives the lane parameter from the lane parameter generation unit,

receives the vehicle state information from the vehicle state information reception unit,

analyzes a time flow-related correlation between the vehicle parameter and the vehicle state information, and

generates an equivalent model based on the analyzed time flow-related correlation;

an erroneous detection determination unit that determines and classifies an erroneous detection of a lane parameter by simultaneously analyzing a case, in which a difference occurs from the equivalent model generated by the equivalent model generation unit, for each of at least one error; and

a torque control factor adjustment unit that adjusts a torque control factor according to a preset torque control factor adjustment setup so as to suppress or reduce the influence of erroneous torque when the erroneous detection determination unit determines the erroneous detection.

2. The lane keeping control system according to claim 1, wherein the lane parameter generation unit generates lane parameters, wherein the lane parameters include coefficients of a third-order equation expressing a relative position between the lane and the vehicle on a two-dimensional plane.

3. The lane keeping control system according to claim 1, wherein the vehicle state information received by the vehicle state information reception unit comprises at least one selected from vehicle longitudinal velocity, vehicle longitudinal acceleration, vehicle yaw rate, steering angle, steering toque, and vehicle transverse acceleration.

4. The lane keeping control system according to claim 1, wherein the equivalent model generation unit generates an equivalent model by calculating the time flow-related correlation between the vehicle parameter and the vehicle state information as an equivalent function from the received lane parameter and the vehicle state information.

5. The lane keeping control system according to claim 1, wherein the erroneous detection determination unit to determines the erroneous detection when a difference value in the equivalent model generated by the equivalent model generation unit exceeds a preset threshold.

6. The lane keeping control system according to claim 5, wherein the erroneous detection determination unit determines the erroneous detection when a difference value in the equivalent model generated by the equivalent model generation unit exceeds a preset threshold for a preset period of time.

7. The lane keeping control system according to claim 1, wherein the error comprises at least one selected from a left lane position error, a right lane position error, a left lane heading angle error, a right lane heading angle error, a both lane heading angle error, a left lane curvature error, a right lane curvature error, and a both lane curvature error.

8. The lane keeping control system according to claim 1, wherein the preset torque control factor adjustment setup by the torque control factor adjustment unit comprises at least one selected from replacement of at least one torque control factor with a different signal, isolation of at least one torque control factor, and adjustment of at least one torque control factor.

9. A vehicle control system comprising:

a vehicle camera that is installed on a vehicle and that collects lane data;

at least one vehicle sensor that is installed on the vehicle and that senses at least one piece of vehicle state information;

a fault detection device that

generates a lane parameter from lane data collected from the vehicle camera,

receives the vehicle state information from the vehicle sensor,

analyzes a time flow-related correlation between the vehicle parameter and the vehicle state information,

generates an equivalent model based on the analyzed time flow-related correlation, and

determines whether an erroneous detection of the lane data is occurred; and

an electronic power steering (EPS) that receives a steering torque signal from the fault detection device according to a situation of erroneous detection of a lane parameter for each error of the fault detection device, reflects the steering torque signal on assist torque regarding a driver of the vehicle, and thereby controls a steering device.

10. The vehicle control system according to claim 9, wherein the fault detection device comprises:

a lane parameter generation unit that generates a relative position between a lane and a vehicle as a lane parameter based on information received from a camera installed on the vehicle;

a vehicle state information reception unit that receives vehicle state information sensed or measured by each vehicle sensor or device installed on the vehicle;

an equivalent model generation unit that

receives the lane parameter from the lane parameter generation unit,

receives the vehicle state information from the vehicle state information reception unit,

analyzes a time flow-related correlation between the vehicle parameter and the vehicle state information, and

generates an equivalent model;

an erroneous detection determination unit that determines and classifies an erroneous detection of a lane parameter by simultaneously analyzing a case, in which a difference occurs from the equivalent model generated by the equivalent model generation unit, for each of at least one error; and

a torque control factor adjustment unit that adjusts a torque control factor according to a preset torque control factor adjustment setup so as to suppress or reduce the influence of erroneous torque when the erroneous detection determination unit determines the erroneous detection.

11. A method for lane data fault detection for lane keeping assist, the method performed by a fault detection device including a lane parameter generation unit, a vehicle state information reception unit, an equivalent model generation unit, an erroneous detection determination unit, and a torque control factor adjustment unit, the method comprising:

generating, by the lane parameter generation unit, a relative position between a lane and a vehicle as a lane parameter based on information received from a camera installed on the vehicle;

receiving, by the vehicle state information reception unit, vehicle state information sensed or measured by each vehicle sensor or device installed on the vehicle;

analyzing, by the equivalent model generation unit, a time flow-related correlation between the received vehicle parameter and the vehicle state information and generating an equivalent model based on the analyzed time flow-related correlation;

determining and classifying, by the erroneous detection determination unit, a situation of erroneous detection of a lane parameter by simultaneously analyzing a case, in which a difference occurs in the generated equivalent model for each of at least one error; and

adjusting, by the torque control factor adjustment unit, a torque control factor according to a preset torque control factor adjustment setup so as to suppress or reduce the influence of erroneous torque when the erroneous detection is determined as a result of the determining.

12. The method according to claim 11, wherein, in the generating a relative position, lane parameters are generated, wherein the lane parameters include coefficients of a third-order equation expressing a relative position between a lane and a vehicle on a two-dimensional plane.

13. The method according to claim 11, wherein the received vehicle state information comprises at least one selected from vehicle longitudinal velocity, vehicle longitudinal acceleration, vehicle yaw rate, steering angle, steering toque, and vehicle transverse acceleration.

14. The method according to claim 11, wherein, in the analyzing, an equivalent model is generated by calculating the time flow-related correlation between the vehicle parameter and the vehicle state information as an equivalent function from the received lane parameter and the vehicle state information.

15. The method according to claim 11, wherein, in the determining and classifying, the erroneous detection is determined when a difference value in the generated equivalent model exceeds a preset threshold.

16. The method according to claim 15, wherein, in the determining and classifying, the erroneous detection is determined when a difference value in the generated equivalent model exceeds a preset threshold for a preset period of time.

17. The method according to claim 11, wherein the error comprises at least one selected from a left lane position error, a right lane position error, a left lane heading angle error, a right lane heading angle error, a both lane heading angle error, a left lane curvature error, a right lane curvature, and a both lane curvature error.

18. The method according to claim 11, wherein the preset torque control factor adjustment setup comprises at least one selected from replacement of at least one torque control factor with a different signal, isolation of at least one torque control factor, and adjustment of at least one torque control factor.

20. A non-transitory computer-readable recording medium on which a program for executing a method according to claim 11 is recorded.

19. A vehicle control method, performed by a vehicle control system including a vehicle camera, at least one vehicle sensor, a fault detection device, and an electronic power steering (EPS), the method comprising:

collecting, by the vehicle camera, lane data from a vehicle camera installed on a vehicle;

sensing, by the at least one vehicle sensor, at least one piece of vehicle state information from at least one vehicle sensor installed on the vehicle;

generating, by the fault detection device, a lane parameter from the lane data collected from the vehicle camera;

receiving, by the fault detection device, the vehicle state information from the vehicle sensor, analyzing a time flow-related correlation between the vehicle parameter and the vehicle state information, and generating an equivalent model based on the analyzed time flow-related correlation;

determining, by the fault detection device, whether an erroneous detection of the lane data is occurred based on the generated equivalent model; and

generating, the fault detection device, a steering torque signal according to a situation of erroneous detection of a lane parameter for each error based on the determining whether the erroneous detection is occurred and providing the EPS with the generated steering torque signal.