KR102258102B1 - Method for controlling steering wheel and System therefor - Google Patents

Abstract

The present invention recognizes an effective parking segment based on the synthesized first image and an obstacle map, and generates an initial parking route based on the effective parking segment, wherein the vehicle is converted to the effective parking segment based on the initial parking route. Detecting a parking line of interest from the second image synthesized when entering, and storing the detected parking line of interest in a map of the parking line of interest, to the currently extracted parking line of interest and the parking line of interest stored in the map of the parking line of interest. Compensating the position error of the vehicle by comparing the rotation angle and movement amount calculated based on the rotation angle and movement amount estimated by the differential odometry, and parking in the effective parking section based on the corrected vehicle position. Regenerating a possible parking path, and controlling a steering wheel of the vehicle based on the initial parking path or the regenerated parking path.

Description

A method for controlling a steering wheel and a system therefor therefor

The present invention relates to a steering wheel control method and an apparatus therefor, and in particular, to a method of controlling a steering wheel using an AVM composite image and a parking assistance system based on a distance sensor, and a system therefor.

Parking is a burden for beginners and elderly drivers, and as they are having a lot of difficulty in parking in a particularly small space, the demand for parking assistance systems is increasing. With such market demand and the recent increase in motor-driven power steering (MDPS)-equipped vehicles, the application of ultrasonic-based parking assistance systems that can reduce the burden on the driver and increase convenience at a low cost is increasing.

As described above, in order to recognize a parking partition space with a conventional ultrasonic sensor, a nearby vehicle or an obstacle next to the parking partition space is required. If not, accuracy in recognizing an effective parking partition space may be lowered.

In addition, the most basic logic in such a conventional parking assistance system is a vehicle position estimator, and a differential odometry method using rear wheel sensor data of a vehicle is widely used. The accuracy of recognizing the effective parking lot space may be lowered due to the uncertainty of.

The cause of an error that causes an error in estimating the attitude of the odometry is a systematic error, which is a decisive cause in which the error cause is inherent in the vehicle. The causes of systematic error include misalignment of two wheel diameters, misalignment of the wheel, and kinematic modeling error.

On the other hand, using a means for detecting an obstacle and a means for detecting an image-based parking division line, based on the obstacle map generated by the obstacle detection means, an ROI is selected from an image provided by the image-based means, and a parking division line is detected. The device is described in detail in European patent EP2377728. In such a conventional technique, when the image-based parking division line detection fails, a method of determining an effective parking division based on an obstacle map is described in detail. When is large, the effective parking zone accuracy and recognition rate may be deteriorated.

If the vehicle position estimation error increases due to the vehicle specifications and the uncertainty of the road surface condition, the error in the parking alignment may increase during the parking process, but the prior art does not provide a solution to the advantage.

The present invention has been devised to solve the above problems, and accurately predicts vehicle movement through detection/matching of parking line guide lines and parking division corners from a synthesized image of an AVM system, and a parking division map for effective parking division recognition. And by accurately building an obstacle map, the effective parking zone recognition performance is improved, and the parking control performance is improved so that the vehicle is parked at a desired location by accurately estimating the car position through detection/matching of the parking line of interest even in the process of parking in the effective parking zone In order to improve parking control in the future, we intend to accurately construct a map of the line of interest.

In addition, by using the parking line guide line and parking lot corner information detected in the process of recognizing the parking lot, correct parking lot maps and obstacle maps by correcting vehicle position errors, and parking lot information of interest in the process of parking in an effective parking lot. It is intended to minimize the error in parking alignment through real-time parking path regeneration while performing vehicle position error correction using.

The steering wheel control method according to an embodiment of the present invention includes the steps of recognizing an effective parking segment based on the synthesized first image and an obstacle map, and generating an initial parking route based on the effective parking segment, the initial parking route. Based on the step of detecting a parking line of interest from the synthesized second image when the vehicle enters the effective parking section, and storing the detected parking line of interest in a map of the parking line of interest, the currently extracted parking line of interest and the Compensating the position error of the vehicle by comparing the rotation angle and movement amount calculated based on the parking line of interest stored in the parking line of interest map and the rotation angle and movement amount estimated by differential odometry, and the corrected vehicle position And regenerating a parking path capable of parking in the effective parking section based on and controlling a steering wheel of the vehicle based on the initial parking path or the regenerated parking path.

In an embodiment, the storing of the detected parking line of interest in a map of the parking line of interest detects a parking line of interest including a left parking partition line, a right parking partition line, and a parking line guide line based on the second image. And updating the detected parking line of interest to the map of the parking line of interest.

In an embodiment, in the detecting of the parking line of interest, the second image is searched in a horizontal direction to detect a region having a relatively high gradient, and two candidate points having a parking line brightness pattern characteristic are extracted, A first step of defining a representative line component for the two candidate points, and a region having a relatively high gradient by searching the second image in a vertical direction, and two candidate points having a parking line brightness pattern characteristic And a second step of extracting and defining a representative line component for the two candidate points, and the left parking partition line, the right parking partition line, the left parking partition line, and the The parking line of interest including the parking line guide line perpendicular to the right parking partition line may be detected.

In an embodiment, the parking line brightness pattern characteristic may include a bright area having a predetermined size between a dark line and a dark line.

The steering wheel control system according to another embodiment of the present invention includes a first image synthesized from a plurality of camera images installed in the vehicle before the vehicle enters the effective parking compartment, and when the vehicle enters the effective parking compartment. An AVM (Around View Monitor) system that provides a second image synthesized from a plurality of camera images installed in the vehicle and a parking line of interest detected from the second image, and a distance detection unit that generates an obstacle map based on an ultrasonic sensor , And recognizes the effective parking segment based on the first image and the obstacle map, generates an initial parking route based on the effective parking segment, and based on the current parking line of interest and the parking line of interest stored in the map of the parking line of interest. By comparing the calculated rotation angle and movement amount with the rotation angle and movement amount estimated by differential odometry, correcting the position error of the vehicle, and parking available in the effective parking compartment based on the corrected position of the vehicle And a steering wheel controller configured to regenerate a path and control a steering wheel of the vehicle based on the initial parking path or the regenerated parking path.

In another embodiment, the parking line of interest may include a left parking division line, a right parking division line, and a parking line guide line, and the parking line guide line may be perpendicular to the left parking division line and the right parking division line.

In another embodiment, the AVM system detects a region having a relatively high gradient by searching for the second image in a horizontal direction and a vertical direction, respectively, and extracts two candidate points having a parking line brightness pattern characteristic, The parking line of interest including the left parking division line, the right parking division line, and the parking line guide line may be detected based on the two candidate points.

According to various embodiments of the present invention, by accurately constructing a parking segment map and an obstacle map for recognizing an effective parking segment from an AVM image, an effective parking segment recognition performance is improved, and a parking line of interest is detected in the process of parking in an effective parking segment. There is an effect of improving the parking control performance so that the vehicle is parked at a desired location by accurately estimating the location of the vehicle through matching.

In addition, there is an effect of minimizing a parking alignment error through real-time parking path regeneration while performing vehicle position error correction using information on a parking line of interest in the process of parking in an effective parking section.

1 is a flowchart illustrating a process of recognizing a parking zone according to an exemplary embodiment of the present invention.
2 is a flowchart illustrating a process of parking a vehicle based on an effective parking zone according to an exemplary embodiment of the present invention.
3 is a flowchart illustrating a process of recognizing an effective parking segment according to another exemplary embodiment of the present invention.
4 is a block diagram illustrating a process of extracting a candidate point having a parking line brightness pattern characteristic in order to detect the parking line guide line shown in FIG. 1 according to an exemplary embodiment of the present invention.
5 is a block diagram illustrating a process of using a line fitting algorithm to detect the parking line guide line shown in FIG. 1 according to an exemplary embodiment of the present invention.
6 is a block diagram illustrating a method of detecting a corner of a parking division shown in FIGS. 1 and 3 according to an exemplary embodiment of the present invention.
7 is a block diagram showing a test result of a method for detecting a corner of a parking division according to an exemplary embodiment of the present invention.
8 is a block diagram illustrating a method of correcting a parking line guide line error according to an exemplary embodiment of the present invention.
9A to 9C are block diagrams and flowcharts illustrating a method of correcting a corner error in a parking division according to an exemplary embodiment of the present invention.
10A to 10B are block diagrams illustrating a method of detecting a parking line of interest according to an exemplary embodiment of the present invention.
11 is a block diagram illustrating a method of detecting a parking reference line according to an exemplary embodiment of the present invention.
12 is a block diagram illustrating a method of detecting a parking line guide line for detecting a parking line of interest according to a preferred embodiment of the present invention.
13 is a block diagram illustrating a method of correcting a position of a parked vehicle using a parking line of interest according to a preferred embodiment of the present invention.
14 is a block diagram showing a system for parking a vehicle based on an effective parking partition according to an exemplary embodiment of the present invention.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but will be implemented in various forms different from each other, and only these embodiments make the disclosure of the present invention complete, and common knowledge in the technical field to which the present invention pertains. It is provided to completely inform the scope of the invention to the possessor, and the invention is only defined by the scope of the claims. Meanwhile, terms used in the present specification are for explaining embodiments and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless specifically stated in the phrase. As used in the specification, "comprises" and/or "comprising" refers to the presence of one or more other components, steps, operations and/or elements in which the recited component, step, operation and/or element is Or does not preclude additions.

1 is a flowchart illustrating a process of recognizing a parking zone according to an exemplary embodiment of the present invention.

Referring to FIG. 1, an AVM (Around View Monitoring) system receiving a plurality of camera images as inputs detects a parking partition corner and a parking line guide line from a first composite image of the images. (S10, S11) The parking line guide line indicates a boundary line of a portion where the vehicle is to enter for the first time. The parking partition corner includes at least two corners adjacent to the parking line guide line.

A parking segment map is generated based on the detected parking line guide line and at least two parking segment corners, and then updated based on a second composite image continuously generated. (S16) Additionally, the rotation angle and/or movement amount of the vehicle may be calculated from the first and second synthesis images, and the calculated rotation angle and/or movement amount is estimated based on an odometry technique. The vehicle position error can be corrected by matching the vehicle position. Such position correction may be reflected when the parking zone map is updated. This position error correction method will be described in detail later in connection with FIGS. 8 to 9C.

Meanwhile, the ultrasonic sensor generates and updates an obstacle map based on a distance detection function while detecting a parking line guide line and at least two parking division corners from the plurality of camera images. (S12)

Using the updated obstacle map and parking segment map, presence/absence of an obstacle in a parking segment to be parked and an effective parking segment are recognized. (S13)

When the effective parking segment recognition is successful, an initial parking path is generated to park in the effective parking segment based on the recognized obstacle. (S14, S15)

2 is a flowchart illustrating a process of parking a vehicle based on an effective parking zone according to an exemplary embodiment of the present invention.

Referring to FIG. 2, while a vehicle enters the effective parking section based on the initial parking path generated as in FIG. 1, a parking line of interest is detected from a third composite image of images provided from a plurality of cameras. (S20) The parking line of interest includes a left parking division line, a right parking division line, and a parking line guide line perpendicular to the two division lines with the vehicle as a vertical axis in an effective parking area in which the corresponding vehicle is to be parked. The parking line guide line has the same shape as the parking line guide line for the initial parking route.

The rotation angle and movement amount of the vehicle may be calculated from a third synthesis image of the images and a fourth synthesis image synthesized from successive images of the images, and the calculated rotation angle and movement amount may be calculated based on an odometry technique. The vehicle position may be corrected by matching with the estimated vehicle position. (S22) This position correction may be reflected when the map of the parking line of interest is updated. (S23)

In the process of matching the parking line of interest, a method of detecting a parking line of interest and a method of correcting a vehicle position based on the parking line of interest will be described in detail later in connection with FIGS. 10A to 13.

The corrected vehicle position may be reflected in the obstacle map generated in FIG. 1. (S26)

A parking path for parking in an effective parking zone to be parked is regenerated based on a parking line map of interest and an obstacle map reflecting the position error of the vehicle. (S24)

The steering wheel of the vehicle is controlled based on the recreated parking path. (S25)

3 is a flowchart illustrating a process of recognizing an effective parking segment according to another exemplary embodiment of the present invention.

Referring to FIG. 3, in an AVM (Around View Monitoring) system that receives a plurality of camera images as inputs, a parking partition corner and a parking line guide line are detected from a fifth composite image of the images. (S30, S31) The parking line guide line indicates a boundary line of a portion where the vehicle is to enter for the first time. The parking partition corner includes at least two corners adjacent to the parking line guide line.

Thereafter, by matching the parking segment corner between the fifth composite image of the images and the sixth composite image synthesized from the consecutive images of the images (S32), the vehicle position based on the differential odometry according to the vehicle specifications and the road surface condition is determined. Correct the estimation error. (S33)

The process of matching the parking division corner will be described in detail later in connection with FIGS. 9A to 9C.

The parking segment map generated from the fifth composite image is updated based on the detected parking line guide line, at least two parking segment corners, and the corrected vehicle position. (S34)

Meanwhile, while detecting a parking line guide line and at least two parking division corners from the plurality of camera images, an obstacle map is generated based on an ultrasonic sensor, and the obstacle map is updated by reflecting the corrected vehicle position. (S36)

Using the updated obstacle map and parking segment map, presence/absence of an obstacle in a parking segment to be parked and an effective parking segment are recognized. (S35)

When the effective parking segment recognition is successful, an initial parking path is generated to park in the effective parking segment based on the recognized obstacle. (S37, S38)

4 is a block diagram illustrating a process of extracting a candidate point having a parking line brightness pattern characteristic in order to detect the parking line guide line shown in FIG. 1 according to an exemplary embodiment of the present invention.

4, images (a) and (b) to find the target image feature point shown in c (here, for detecting a predetermined area of a predetermined brightness between a dark line and a dark line for finding a parking line guide line) As shown in, horizontal sections to be searched are selected. In the outline image (a), a location with a relatively high gradient is detected by searching in the horizontal direction and defined as a feature point. The candidate points are extracted by combining and generating two candidate points having a parking line brightness pattern characteristic in each interest section. The parking line brightness pattern characteristic indicates when there is a bright area of a predetermined size between a dark line and a dark line.

5 is a block diagram illustrating a process of using a line fitting algorithm to detect the parking line guide line shown in FIG. 1 according to an exemplary embodiment of the present invention.

Referring to FIG. 5, as shown in (a) and (b), a line component is extracted by searching in a 360-degree direction based on a candidate point (left/right) having a parking line brightness pattern in an outline image. A line component passing through a plurality of candidate points, such as the thick red line shown in (c), is recognized as the final parking line guideline.

6 is a block diagram illustrating a method of detecting a corner of a parking division shown in FIGS. 1 and 3 according to an exemplary embodiment of the present invention.

Referring to FIG. 6, corner feature points are extracted using a Harris corner detection algorithm or the like. (a) The Harris corner detection algorithm is implemented in a way that basically there is a window that moves vertically and horizontally in an image, and a corner point is found by analyzing changes in pixel values in the window. If there is no change in the brightness value of the object in the image, there will be no change in the pixel value even if the window is moved vertically and horizontally. However, when moving left and right and encountering the boundary line of an image that exists in the vertical direction, of course, a large change occurs in the pixel value in the window moving in the left-right direction, but there will be no change in the pixel value for the window moving in the vertical direction. Then, if you think that this window moves not only in the left-right direction, but also in the vertical direction, it will obviously pass the point where the change of the pixel value is large while moving in the vertical direction. That is, this is the final corner point. The corner point has the advantage of finding the same position even when the image is rotated.

Parking division corner type (as shown in c, there are types A and B for the L-shaped connection point and the T-shaped connection point) by detecting a corner with a brightness pattern characteristic of the parking line guide line (red color) The parking division corner passing through the solid line) is finally selected. (b)

7 is a block diagram showing a test result of a method for detecting a corner of a parking division according to an exemplary embodiment of the present invention.

8 is a block diagram illustrating a method of correcting a parking line guide line error according to an exemplary embodiment of the present invention.

Referring to FIG. 8, the vehicle rotation angle is estimated by calculating an angle difference between a parking line guide line on a parking segment map detected and stored from a previous AVM image and a parking line guide line detected from a current AVM image. The error of the vehicle rotation angle can be corrected by comparing the estimated host vehicle rotation angle with the rotation angle estimated by differential odometry.

9A to 9C are block diagrams and flowcharts illustrating a method of correcting a corner error in a parking division according to an exemplary embodiment of the present invention.

9A to 9B, the amount of movement of the own vehicle is estimated by calculating a movement amount between a parking segment corner on a parking segment map detected and stored from a previous AVM image and a parking segment corner detected from a current AVM image.

Referring to FIG. 9C, in order to estimate the amount of movement, first, as shown in FIG. 6, the parking division corner types (types A and B for the L-shaped connection point and the T-shaped connection point as shown in c) are Exists), and at least two parking zone corners passing through the parking line guide line (solid red line) are finally selected. (S40)

These finally selected at least two parking division corners are matched to estimate the amount of movement of the own vehicle.

In more detail, rotational components between two consecutive images are removed by rotationally transforming the intersection points of one image (a) using the own vehicle rotation angle estimated in FIG. 8. (S41) The amount of movement between the intersection points is calculated in the image coordinate system. (S42) The calculated movement amount is converted into a real world coordinate system to calculate the own vehicle movement amount. (S43)

The error of the vehicle movement amount (position) may be corrected by comparing the estimated movement amount of the own vehicle with the movement amount estimated by differential odometry.

10A to 10B are block diagrams illustrating a method of detecting a parking line of interest according to an exemplary embodiment of the present invention.

Referring to FIG. 10A, in order to find the target image feature point shown in d (here, for detecting a predetermined area of a predetermined brightness between a dark line and a dark line for finding a parking line guide line to be included in a parking line of interest), an image (a ) And (b), horizontal sections to be searched are selected. In the outline image (b), a location with a relatively high gradient is detected by searching in the horizontal direction and defined as a feature point. The candidate points are extracted by combining and generating two candidate points having a parking line brightness pattern characteristic in each interest section. The parking line brightness pattern characteristic indicates when there is a bright area of a predetermined size between a dark line and a dark line.

Referring to FIG. 10B, as shown in (a) and (b), a line component is extracted by searching in a 360-degree direction based on a candidate point (left/right) having a parking line brightness pattern in an outline image.

After detecting the line component for the left and right feature points, the angle difference is compared to confirm the extraction of the correct line component. (d) As shown in the image, it can be determined that the line component located on the front of the vehicle is not a legitimate line.

The representative line component is defined by using the line component with the minimum line fitting error among the two line components. In addition, as shown in (c) and (d), it is expressed as a line component passing through the center point of the left/right feature point. This is to find the line component passing through the center of the parking line.

11 is a block diagram illustrating a method of detecting a parking reference line according to an exemplary embodiment of the present invention.

Referring to FIG. 11, a left/right parking line of interest region of interest (red rectangle) is defined based on the longitudinal axis of the host vehicle. (a)

In the left/right parking line area of interest, the optimal line component that best reflects the candidate points having the parking line brightness pattern is selected. (b)

If the direction error and spacing of the left/right parking line of interest (yellow solid line) satisfies the parking zone geometry condition, the parking reference line (red dotted line) is calculated. (c)

12 is a block diagram illustrating a method of detecting a parking line guide line for detecting a parking line of interest according to a preferred embodiment of the present invention.

Referring to FIG. 12, except for searching in the vertical direction in the outline image (a), it is the same as the parking line guide line detection method shown in FIG. 10A, and a parking line guide line forming a vertical relationship with the left/right parking division lines is used. To detect.

13 is a block diagram illustrating a method of correcting a position of a parked vehicle using a parking line of interest according to an exemplary embodiment of the present invention.

13, after converting the parking line of interest extracted from the current image into a real-world map coordinate system, calculating the rotation angle and the amount of movement at which the error is minimum with the parking line of interest stored in the map of the parking line of interest, The position error of the vehicle may be corrected by comparing it with the estimated rotation angle and movement amount of the own vehicle.

14 is a block diagram showing a system for parking a vehicle based on an effective parking partition according to an exemplary embodiment of the present invention.

Referring to FIG. 14, the steering wheel system according to an embodiment of the present invention detects a parking partition corner and a parking line guideline from a first image synthesized from a plurality of camera images installed in a vehicle, and determines an initial parking path. Based on an AVM (Around View Monitor) system (10) that detects a parking line of interest from a second image synthesized from a plurality of camera images installed in the vehicle when the vehicle enters the effective parking compartment while referring to it, and an ultrasonic sensor The distance detection unit 20 for generating an obstacle map, the detected parking segment corner and the parking line guideline, and the presence/absence of an obstacle in the parking segment to be parked and the effective parking segment based on the map based on the generated obstacle. And, based on the recognized obstacle and the effective parking section, to create an initial parking path to park in the effective parking section, and based on at least one of the detected parking line guide line, parking section corner, and parking lines of interest To correct the position of the vehicle, regenerate a parking path capable of parking in the effective parking section based on the corrected vehicle position, and a steering wheel of the vehicle based on the initial parking path or the regenerated parking path It is configured to include a steering wheel control unit 30 to control.

For example, the steering wheel control unit 30 calculates the rotation angle and movement amount of the vehicle from the parking lines of interest detected from the first image and the second continuous image of the first image, and the calculated rotation The position of the vehicle may be corrected by matching the angle and the amount of movement with the vehicle position estimated based on the odometry.

The odometry estimating unit 40 for estimating the vehicle position may use rear wheel sensor data of the vehicle.

The parking line of interest may include a left parking division line, a right parking division line, and a parking line guide line perpendicular to the two division lines with the vehicle as a vertical axis.

The embodiments of the present invention described above are not implemented only through an apparatus and a method, but may be implemented through a program that realizes a function corresponding to the configuration of the embodiment of the present invention or a recording medium in which the program is recorded. Implementation can be easily implemented by an expert in the technical field to which the present invention belongs from the description of the above-described embodiment.

The present invention described above is capable of various substitutions, modifications, and changes within the scope of the technical spirit of the present invention to those of ordinary skill in the art. It is not limited by the drawings, and all or part of each of the embodiments may be selectively combined so that various modifications may be made.

10: AVM system
20: distance detection unit
30: steering wheel control unit
40: odometry estimation unit

Claims (7)

Recognizing an effective parking segment based on the synthesized first image and an obstacle map, and generating an initial parking route based on the effective parking segment;
Detecting a parking line of interest from a second image synthesized when a vehicle enters the effective parking section based on the initial parking route, and storing the detected parking line of interest in a map of the parking line of interest;
Compensating the position error of the vehicle by comparing the rotation angle and movement amount calculated based on the currently extracted parking line of interest and the parking line of interest stored in the map of the parking lot of interest, and the rotation angle and movement amount estimated by differential odometry. ; And
Regenerating a parking path capable of parking in the effective parking section based on the corrected position of the vehicle, and controlling a steering wheel of the vehicle based on the initial parking path or the regenerated parking path;
Steering wheel control method comprising a.
The method according to claim 1,
The step of storing the detected parking line of interest in a map of the parking line of interest,
Detecting a parking line of interest including a left parking partition line, a right parking partition line, and a parking line guide line based on the second image, and
And updating the detected parking line of interest to the map of the parking line of interest.
The method according to claim 2,
The step of detecting the parking line of interest,
A first for searching the second image in a horizontal direction to detect a region having a relatively high gradient, extracting two candidate points having a parking line brightness pattern characteristic, and defining a representative line component for the two candidate points Step, and
A second image that searches the second image in a vertical direction to detect a region having a relatively high gradient, extracts two candidate points having a parking line brightness pattern characteristic, and defines a representative line component for the two candidate points. Includes steps,
Detecting the parking line of interest including the left parking division line, the right parking division line, and the parking line guide line perpendicular to the left parking division line and the right parking division line through the first and second steps. Steering wheel control method, characterized in that.
The method of claim 3,
The parking line brightness pattern characteristic,
A method of controlling a steering wheel, comprising a bright area having a predetermined size between the dark line and the dark line. A first image synthesized from a plurality of camera images installed in the vehicle before the vehicle enters the effective parking segment, and a second image synthesized from a plurality of camera images installed in the vehicle when the vehicle enters the effective parking segment. An AVM (Around View Monitor) system that provides an image and a parking line of interest detected from the second image;
A distance detector for generating an obstacle map based on the ultrasonic sensor; And
Recognizes the effective parking segment based on the first image and the obstacle map, generates an initial parking route based on the effective parking segment, and calculates based on the current parking line of interest and the parking line of interest stored in the map of the parking line of interest. Compensating the position error of the vehicle by comparing the calculated rotation angle and movement amount with the rotation angle and movement amount estimated by the differential odometry, and based on the corrected position of the vehicle, a parking path for parking in the effective parking compartment is determined. A steering wheel controller configured to regenerate and control a steering wheel of the vehicle based on the initial parking path or the regenerated parking path;
Steering wheel control system comprising a.
The method of claim 5,
The parking line of interest,
Including the left parking division line, the right parking division line and the parking line guide line,
The parking line guide line,
The steering wheel control system, characterized in that perpendicular to the left parking partition line and the right parking partition line.
The method of claim 6,
The AVM system,
The second image is searched in a horizontal direction and a vertical direction, respectively, to detect a region having a relatively high gradient, extract two candidate points having a parking line brightness pattern characteristic, and the left parking based on the two candidate points. The steering wheel control system, characterized in that to detect the parking line of interest including the division line, the right parking division line, and the parking line guide line.

Images