Method and apparatus for positioning using AVPS markers

Abstract

The invention provides a method and a device for positioning a vehicle by using an AVPS mark. According to at least one embodiment, the present disclosure provides a method for performing positioning by a vehicle including a plurality of cameras, including identifying a corner point of a coded mark from an image acquired by each camera, calculating a position, a width, and an angle of the coded mark by using coordinates of the corner point, filtering the coded mark, and performing positioning by using the filtered coded mark. In one embodiment, a method for measuring a position of a vehicle includes encoding a mark in a parking facility, calculating a final weight of the encoded mark, generating a position and an angle of a local encoded mark, rotationally transforming the position of the local encoded mark by using information about the mark in the parking facility and the angle of the local encoded mark, and measuring the position of the vehicle by using the position of the local encoded mark and the final weight.

Description

[0001] Cross - reference to related applications

[0002] This application claims the benefit of Korean Patent Application No. 10 - 2023 - 0181560, filed on December 14, 2023, the entire contents of which are incorporated herein by reference. Technical field

[0003] The present disclosure relates to a method and apparatus for positioning using AVPS markers, and more particularly, to a method and apparatus for improving the positioning accuracy of an autonomous driving vehicle using AVPS markers. Background art

[0004] The statements herein merely provide background information related to the present disclosure and do not necessarily constitute prior art.

[0005] For convenient parking, an automated valet parking system (AVPS) is being developed. The AVPS automatically operates a vehicle so that the vehicle moves to an empty parking space and parks when the driver gets off at the drop - off area of the parking lot. In addition, upon the driver's request, the AVPS automatically moves the parked vehicle to the pick - up area to allow the driver to board the vehicle.

[0006] For a safe and reliable AVPS, level 4 or higher autonomous driving is required. The AVPS must not only identify other objects to prevent collisions, but also determine empty parking spaces and driving routes, and automatically move, park, and retrieve the vehicle. For this, positioning technology for estimating the current position of a target vehicle is important.

[0007] The AVPS employs vision - based positioning using a high - definition map and a camera for positioning. The AVPS uses coded markers defined in ISO 23374 for positioning. The coded marker is a fiducial marker that can be installed in a parking facility and recognized by a vehicle. The fiducial marker is an artificial marker that appears in a generated image as an object placed in the field of view of an imaging system and serves as a reference point or a measurement point. The high - definition map can include information such as the ID, location, orientation, etc. of the coded markers installed in the parking facility.

[0008] Conventional methods for detecting and recognizing coded markers have problems, including low accuracy in detecting coded markers due to the distance between the camera and the marker, distortion of the coded markers in the image, etc., and difficulty in accurately estimating the distance to the marker. Therefore, a positioning method is needed that accurately estimates the distance between the camera and the marker by considering distortion and accurately measures the position of the vehicle using the estimated distance. Summary of the invention

[0009] Embodiments of the present disclosure provide a method and apparatus for accurately estimating the distance between a camera and a marker by considering distortion and positioning a vehicle based on the estimated distance.

[0010] Embodiments of the present disclosure are not limited to the above embodiments, and those of ordinary skill in the art will clearly understand other embodiments not mentioned above through the following description.

[0011] At least one embodiment of the present disclosure provides a method for vehicle positioning performed by a vehicle including a plurality of cameras, including: identifying corner points of an encoded marker from images obtained from each camera; calculating coordinates of the corner points, and calculating the position, width, and angle of the encoded marker by using the coordinates of the corner points; filtering the encoded marker based on the position and width of the encoded marker; calculating a final weight of the encoded marker; transforming the angle and position of the encoded marker by using the coordinates of one camera as vehicle coordinates to generate the position and angle of a local encoded marker; rotating and transforming the position of the local encoded marker by using information about the angle of the marker in a parking facility and the local encoded marker; and measuring the position of the vehicle by using the position of the local encoded marker and the final weight.

[0012] Another embodiment of the present disclosure provides an apparatus for vehicle positioning, including a memory for storing instructions; and at least one processor, wherein, by executing the instructions, the at least one processor identifies corner points of an encoded marker from images obtained from each camera, calculates coordinates of the corner points, calculates the position, width, and angle of the encoded marker by using the coordinates of the corner points, filters the encoded marker based on the position and width of the encoded marker, calculates a final weight of the encoded marker, transforms the angle and position of the encoded marker by using the coordinates of one camera as vehicle coordinates to generate the position and angle of a local encoded marker, rotates and transforms the position of the local encoded marker by using information about the angle of the marker in a parking facility and the local encoded marker, and measures the position of the vehicle by using the position of the local encoded marker and the final weight.

[0013] According to an embodiment of the present disclosure, accurate positioning can be performed by providing a method in which: an encoded marker is identified by using a plurality of cameras, a weight is generated by considering the distance and distortion between the encoded marker and the camera, and the encoded markers identified by the plurality of cameras are transformed into one coordinate system.

[0014] The effects of the present disclosure are not limited to the above, and those skilled in the art will be able to clearly understand other effects not mentioned herein through the following description. Description of the Drawings

[0015] Figure 1is a block diagram schematically showing a vehicle positioning device according to an embodiment of the present disclosure.

[0016] Figure 2 is a diagram for explaining the calculation of the position of the first corner point of a coded marker according to an embodiment of the present disclosure.

[0017] Figure 3 is a diagram for explaining the calculation of the angle and distance of a coded marker according to an embodiment of the present disclosure.

[0018] Figure 4 is a flowchart showing a vehicle positioning method according to an embodiment of the present disclosure. Detailed Description of the Invention

[0019] Hereinafter, some embodiments of the present disclosure will be described in detail with reference to the accompanying illustrative drawings. In the following description, the same reference numerals preferably denote the same elements even if the elements are shown in different drawings. In addition, in the description of some embodiments below, for the purpose of clarity and conciseness, detailed descriptions of related known components and functions will be omitted when it is considered that they obscure the subject matter of the embodiments of the present disclosure.

[0020] When various ordinal numbers or letter codes (such as first, second, i), ii), a), b), etc.) are used as prefixes, they are only used to distinguish one component from another, rather than implying or suggesting the essence, order or sequence of the components. In this specification, when a part "includes" or "comprises" a component, this part means that other components are also included, and other components are not excluded unless there is a clear contrary statement. Terms (such as "unit", "module", etc.) refer to one or more units for processing at least one function or operation, which can be implemented by hardware, software or a combination thereof.

[0021] The description of the embodiments of the present disclosure presented below with reference to the drawings is intended to describe the exemplary embodiments of the present disclosure, and is not intended to represent the only embodiments in which the technical idea of the present disclosure can be practiced.

[0022] The present invention relates to a technique for improving the accuracy of detecting a coded marker for vehicle positioning in an automated valet parking system (AVPS).

[0023] Figure 1 is a block diagram schematically showing a vehicle positioning device according to an embodiment of the present disclosure.

[0024] Referring to Figure 1 , the vehicle positioning device may include all or part of a marker recognition unit 100, a marker unification unit 110, an odometry unit 120, and a positioning unit 130.

[0025] The marker recognition unit 100 obtains images from multiple cameras. For example, the multiple cameras can be cameras respectively used to capture the front, left, right, and rear of the vehicle. The marker recognition unit 100 obtains, for example, the images of the front, left, right, and rear of the captured vehicle, and identifies the coded markers by using these images. According to an embodiment of the present disclosure, one image may include at least one coded marker.

[0026] The marker recognition unit 100 identifies the corner points of the coded marker. One coded marker has four corner points. The marker recognition unit 100 identifies the ID of the coded marker by using the identified corner points.

[0027] The ranging unit 120 calculates the relative change in the vehicle position. The ranging unit 120 obtains the specifications of the vehicle. The specifications of the vehicle may include, for example, the wheelbase and wheel size of the vehicle. The ranging unit 120 periodically receives the driving information of the vehicle. The driving information includes, for example, the steering angle and wheel pulse. The ranging unit 120 periodically calculates the ranging of the vehicle by using the specifications of the vehicle and the driving information of the vehicle. According to an embodiment of the present disclosure, the ranging unit 120 may calculate the ranging of the vehicle whenever it receives the driving information. According to an embodiment of the present disclosure, the ranging includes the relative longitudinal and lateral distances traveled by the vehicle. The ranging unit 120 periodically stores the calculated ranging readings in a ranging buffer.

[0028] The ranging unit 120 calculates the longitudinal and lateral distances traveled by the vehicle over a period of time by using the ranging readings stored in the ranging buffer. For example, if fifteen ranging readings are calculated within one second, the ranging unit 120 may periodically calculate the traveled distance for about 66 ms.

[0029] The ranging unit 120 measures the delay from the time point when the image is obtained from the camera to the time point when the vehicle performs positioning by using the coded marker identified in the image. The ranging unit 120 calculates the distance traveled by the vehicle during the delay, i.e., the delay distance, by using the ranging.

[0030] The marker unification unit 110 may include all or part of an attitude estimation unit 112, a filtering unit 114, a coordinate conversion unit 116, and a marker tracking unit 118.

[0031] The attitude estimation unit 112 estimates the position and angle of each identified coded marker. Here, the position of each coded marker is the position relative to the origin on the ground perpendicular to the position of the camera.

[0032] Hereinafter, reference will be made to Figure 2 and Figure 3 to describe the method by which the attitude estimation unit 112 estimates the position and angle of each coded marker. Figure 2 andFigure 3 The coordinates in

[0033] Figure 2 are relative to the origin, which is the number of feet perpendicular from the camera's position to the ground.

[0034] Reference Figure 2 , a coded marker has four corner points. The coded marker is square. According to an embodiment of the present disclosure, the distance between the nearest corner points can be 300 mm. The distance between the nearest corner points can be calculated from the size of the preset coded marker.

[0035] The pose estimation unit 112 calculates the coordinates of the first corner point of the coded marker by using the distance from the camera to the standardized image plane, the height of the camera from the ground, and the distance between the nearest corner points. The first corner point refers to the corner point located at the upper left end of the coded marker. The four corner points of a coded marker include the first corner point and the second, third, and fourth corner points placed clockwise from the first corner point. The standardized image plane is a plane at a preset distance from the focus of the camera. According to an embodiment of the present disclosure, the standardized image plane can be a plane parallel to the coded marker. The pose estimation unit 112 obtains the coordinates of each corner point on the standardized image plane.

[0036] The x - coordinate and z - coordinate of the first corner point on the standardized image plane are represented by Pw1 x and Pw1 z respectively. The x - coordinate and z - coordinate of the fourth corner point on the standardized image plane are represented by Pw4 x and Pw4 z respectively.

[0037] The first corner point and the fourth corner point of the actual coded marker have the same x and y coordinates. Let the x - coordinate of the first corner point of the actual coded marker be long1, and the z - coordinate of the first corner point of the actual coded marker be height1. Then, the x - coordinate of the fourth corner point of the actual coded marker is long1, and the z - coordinate of the fourth corner point of the actual coded marker is height4. Here, height4 is a value obtained by subtracting 300 mm from height1.

[0038] The relationship between the height Tw z of the camera from the ground, the coordinates of the first corner point on the standardized image plane, and the coordinates of the first corner point of the actual coded marker is given by Equation 1:

[0039] (Equation 1)

[0040]

[0041] where Tw z is the height of the camera from the ground.

[0042] The relationship among the height of the camera, the coordinates of the fourth corner point on the standardized image plane, and the coordinates of the fourth corner point of the actual coding mark is given by Equation 2:

[0043] (Equation 2)

[0044]

[0045] Equation 1 and Equation 2 can be summarized as Equation 3.

[0046] (Equation 3)

[0047]

[0048] The y - coordinates of the first and fourth corner points of the coding mark are given by Equation 4:

[0049] (Equation 4)

[0050]

[0051] Since the relationship between the second and third corner points is the same as that between the first and fourth corner points, the coordinates of the second corner point can be calculated in the same way as the coordinates of the first corner point. In other words, the coordinates of the first to fourth corner points can be calculated by using Equations 1 to 4.

[0052] Figure 3 is a diagram for explaining the calculation of the angle and distance of the coding mark according to an embodiment of the present disclosure.

[0053] Refer to Figure 3 , the distance between the first and second corner points is 300 mm. The first and second corner points have the same height, that is, the same z - coordinate.

[0054] The angle A of the coding mark can be calculated by Equation 5:

[0055] (Equation 5)

[0056]

[0057] where lat1 is the y - coordinate of the first corner point and lat2 is the y - coordinate of the second corner point.

[0058] The distance B of the coding mark refers to the distance between the position of the camera and the center point of the coding mark. That is, the distance B of the coding mark can be calculated by Equation 6.

[0059] (Equation 6)

[0060]

[0061] where long c is the x - coordinate of the center point, lat c is the y - coordinate of the center point, and height c is the z - coordinate of the center point. The coordinates of the center point of the coded marker can be calculated by using the coordinates of the first corner point to the fourth corner point. Hereinafter, the coordinates of the center point of the coded marker are referred to as the position of the coded marker.

[0062] The pose estimation unit 112 can calculate the width of the coded marker. The width of the coded marker can be the distance between the first corner point and the second corner point, which is calculated by using the coordinates from the first corner point to the second corner point.

[0063] According to an embodiment of the present disclosure, the vehicle positioning device may include at least one pose estimation unit 112. One pose estimation unit 112 may correspond to one camera. In other words, one pose estimation unit 112 can receive an image from one camera and calculate the angle, distance, position, and width of the coded marker recognized in the received image.

[0064] The filtering unit 114 performs filtering by using the position and width of the recognized coded marker.

[0065] The filtering unit 114 deletes the recognition information of the coded marker unless the position of the coded marker exists in the region of interest (ROI). If the width of the coded marker is different from a specific preset value, the filtering unit 114 deletes the recognition information of the coded marker. In other words, if the coded marker exists outside the ROI or its width is different from the specific preset value, the filtering unit 114 deletes the recognition information of the coded marker. For example, the preset value can be 300 mm.

[0066] The filtering unit 114 calculates the weight of the coded marker. The weight of the coded marker includes at least one of a distance weight and a distortion weight. The distance weight is the weight relative to the distance between the camera and the coded marker. The distortion weight is the weight relative to the degree of distortion of the coded marker in the image.

[0067] The distance weight is a value obtained by dividing the distance between a certain coded marker and the camera by the sum of the distances between all recognized coded markers and the camera. That is, the distance weight is the ratio of the distance of a certain coded marker to the sum of the distances of all recognized coded markers. The greater the distance of the coded marker, the higher the weight. The distance weight of the k - th coded marker is calculated by Equation 7:

[0068] (Equation 7)

[0069]

[0070] where n is the number of recognized coded markers, and d k is the distance between the k-th recognized coded marker and the camera.

[0071] The filtering unit 114 calculates a distortion weight by using the distance between the corner points and the center point of the coded marker. The distortion weight of the k-th coded marker is calculated by Equation 8:

[0072] (Equation 8)

[0073]

[0074] where d k 1 is the distance between the first corner point of the k-th recognized coded marker and the center point of the k-th recognized coded marker, and d k 2 is the distance between the second corner point of the k-th recognized coded marker and the center point of the k-th recognized coded marker, and d k 3 is the distance between the third corner point of the k-th recognized coded marker and the center point of the k-th recognized coded marker, and d k 4 is the distance between the fourth corner point of the k-th recognized coded identifier and the center point of the k-th recognized coded identifier, and d k avg is d k 1, d k 2, d k 3, d k 4's average value.

[0075] The filtering unit 114 generates a final weight. The final weight of the k-th coded marker is the sum of the distance weight of the k-th coded marker and the distortion weight of the k-th coded marker.

[0076] The filtering unit 114 deletes the recognition information of the coded markers except for the coded marker with the highest final weight among the coded markers with the same ID recognized by different cameras.

[0077] The coordinate conversion unit 116 converts the coordinates of the coded marker into vehicle coordinates.

[0078] The vehicle coordinates can be the coordinates of one of the cameras. For example, the coordinates of the front camera can be determined as the vehicle coordinates, and the coordinates of the coded markers recognized by the rear, left, and right cameras can be converted into vehicle coordinates. The camera used as a reference for the vehicle coordinates is called the reference camera.

[0079] The coordinate conversion unit 116 can convert the coordinates of the coded marker by using the position of the coded marker, the angle of the coded marker, and the distance between the cameras. The front, rear, left, and right cameras are respectively placed at the front, rear, left, and right of the vehicle. The cameras are placed perpendicular to adjacent cameras. The vertical distance and the horizontal distance between the cameras are values included in the specifications of the vehicle. The coordinate conversion unit 116 converts the coordinates of the coded marker recognized by each camera by using the angle between each camera and a reference camera, and the vertical distance and the horizontal distance between the cameras.

[0080] The marker tracking unit 118 predicts the position and angle of the coded marker by using the ranging readings from the ranging unit 120 stored in the ranging buffer. The marker tracking unit 118 can periodically predict the position and angle of the coded marker. For example, the marker tracking unit 118 can predict the position and angle of the coded marker at the same period as the ranging unit 120. The marker tracking unit 118 generates a predicted position and a predicted angle by predicting the position and angle of the coded marker. The marker tracking unit 118 can generate a predicted position and a predicted angle for each recognized coded marker.

[0081] The marker tracking unit 118 correlates the position and angle of the actually recognized coded marker with the predicted position and predicted angle. The marker tracking unit 118 corrects the predicted position and predicted angle by correlating the recognized coded marker information (such as position, angle, final weight, and ID) with the predicted position and predicted angle.

[0082] The marker tracking unit 118 stores the corrected position and angle of the coded marker, and uses the stored position and angle of the coded marker to generate the predicted position and predicted angle for the next frame.

[0083] The positioning unit 130 estimates the global position of the vehicle. The vehicle and the vehicle positioning device are in the same global position. The positioning unit 130 can receive information about the markers in the parking facility from the parking facility control center, and estimate the global position by using the information about the markers in the parking facility.

[0084] The positioning unit 130 searches for a coded marker (hereinafter referred to as "global coded marker") having the same ID as the coded marker (hereinafter referred to as "local coded marker") recognized by the vehicle positioning device among the coded markers included in the information about the markers in the parking facility. The positioning unit 130 compares the angles of the global coded marker and the local coded marker having the same ID, and calculates the difference. The positioning unit 130 performs a rotation transformation on the position of the local coded marker by the calculated angle difference.

[0085] The positioning unit 130 determines whether the rotation transformation position of the coded marker is within the drivable area. In addition, the positioning unit 130 determines whether the difference between the position of the coded marker after rotation transformation and the previous position of the vehicle is greater than or equal to a certain preset value. If the position of the coded marker after rotation transformation exists outside the drivable area or the difference between the position of the rotated marker and the previous position of the vehicle is greater than or equal to a certain preset value, the positioning unit 130 deletes the identification information about the coded marker.

[0086] The positioning unit 130 corrects the position of the vehicle by using the vehicle position and the ranging of the previous frame, and generates the final vehicle position by adding the delay distance to the corrected vehicle position.

[0087] Figure 4 It is a flowchart showing a vehicle positioning method according to an embodiment of the present disclosure.

[0088] Refer to Figure 4 , the marker recognition unit 100 obtains images containing coded markers from multiple cameras. The marker recognition unit 100 recognizes the coded markers in the images. The marker recognition unit 100 recognizes the four corner points on each coded marker and the ID of each coded marker (S400).

[0089] The attitude estimation unit 112 calculates the position, distance, width, and angle of the coded marker by using the coordinates of the corner points of the coded marker recognized by the marker recognition unit 100 on the standardized image plane (S410).

[0090] The filtering unit 114 filters the identification information of the coded marker by using the position and width of the coded marker (S420). The filtering unit 114 can filter out the identification information of the coded marker located outside the ROI. The filtering unit 114 can filter out the identification information of the coded marker whose width is different from the preset value. Here, the preset value can be the size of the coded marker.

[0091] The filtering unit 114 calculates the distance weight and the distortion weight for each coded marker, and generates the final weight by adding the distance weight and the distortion weight (S430). The distance weight is the weight regarding the distance between the coded marker and the camera. The distortion weight is the weight regarding the degree of distortion of the coded marker in the image.

[0092] The coordinate conversion unit 116 determines the reference camera and determines the coordinates of the reference camera as the vehicle coordinates. The coordinate conversion unit 116 converts the position and angle of the coded marker obtained from cameras other than the reference camera into vehicle coordinates (S440). The coordinate conversion unit 116 converts the position and angle of the coded marker into vehicle coordinates by using the angle between the cameras and the vertical and horizontal distances between the cameras.

[0093] The marker tracking unit 118 estimates the position and angle of the coded marker in the next frame by using the ranging readings calculated by the ranging unit 120 and the previously acquired position and angle of the coded marker. The marker tracking unit 118 corrects the position and angle of the coded marker by associating the position and angle of the identified coded marker with the estimated position and angle (S450).

[0094] The positioning unit 130 converts the position and angle of the coded marker into the position and angle on the global coordinate system (S460). The positioning unit 130 receives information about the markers in the parking facility. The positioning unit 130 searches for the global coded marker having the same ID as the identified coded marker (i.e., the local coded marker) in the information about the markers in the parking facility, and performs a rotation transformation on the position of the local coded marker by using the angular difference between the local coded marker and the global coded marker.

[0095] The positioning unit 130 measures the current position of the vehicle. The positioning unit 130 measures the position of the vehicle by using the positions of a plurality of coded markers and the final weights of the coded markers (S470). The positioning unit 130 corrects the position of the vehicle by using the position, ranging, and delay distance of the vehicle in the previous frame.

[0096] Each component of the device or method according to an embodiment of the present disclosure may be implemented in hardware, software, or a combination of hardware and software. In addition, the functions of each component may be implemented in software, and a microprocessor may be used to execute the software functions corresponding to each component.

[0097] Various implementations of the systems and techniques described herein may include digital electronic circuits, integrated circuits, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), computer hardware, firmware, software, and / or combinations thereof. These different implementations may include one or more computer programs executable on a programmable system. The programmable system includes at least one programmable processor (which may be a dedicated processor or a general-purpose processor), which is combined to receive data and instructions from a storage system, at least one input device, and at least one output device, and to send data and instructions to them. A computer program (also referred to as a program, software, software application, or code) includes instructions for a programmable processor and is stored in a "computer-readable storage medium".

[0098] A computer-readable storage medium includes various storage devices that can store data readable by a computer system. The computer-readable storage medium may include non-volatile or non-transitory media (such as Read-Only Memory (ROM), Compact Disc Read-Only Memory (CD-ROM), magnetic tapes, floppy disks, memory cards, hard disks, magneto-optical discs, and storage devices), and may also include transitory media (such as data transmission media). In addition, the computer-readable storage medium may be distributed among computer systems connected by a network, and the computer-readable code may be stored and executed in a distributed manner.

[0099] In the flowcharts of this specification, each process is described as occurring sequentially, but this is merely an example of the technology of the embodiments of the present disclosure. In other words, those of ordinary skill in the art to which the embodiments of the present disclosure pertain can make various modifications and variations by changing the order described in the flowcharts of this specification or by undergoing one or more processes in parallel within the basic features of an embodiment of the present disclosure. Therefore, the flowcharts of this specification are not limited to the time-series order.

[0100] Although the exemplary embodiments of the present disclosure have been described for illustrative purposes, those skilled in the art will understand that various modifications, additions, and substitutions can be made without departing from the spirit and scope of the claimed invention. Therefore, the exemplary embodiments of the present disclosure have been described for the sake of brevity and clarity. The scope of the technical idea of the embodiments of the present disclosure is not limited by the illustrations. Therefore, those of ordinary skill in the art will understand that the scope of the claimed invention is not limited by the embodiments explicitly described above, but rather by the technical solutions and their equivalents.

Claims

1. A method for performing positioning by a vehicle including a plurality of cameras, comprising: identifying corner points of the coded mark from images acquired by each of the plurality of cameras; Calculating the coordinates of the corner points, and calculating the position, width and angle of the coding mark by using the coordinates of the corner points; filtering the coded marks based on the positions and widths of the coded marks; Calculating the final weight of the encoding mark; Converting the angle and position of the coded marker by using the coordinates of one of the plurality of cameras as vehicle coordinates to generate a position and angle of a local coded marker; rotationally transforming the position of the locally encoded marker by using information about the marker in the parking facility and the angle of the locally encoded marker; as well as The position of the vehicle is measured by using the position of the locally encoded marker and the final weight.

2. The method of claim 1 further comprising periodically calculating the relative distance traveled by the vehicle.

3. The method according to claim 2, further comprising: By using the relative distance traveled by the vehicle and the position and angle of the local coding marker of the previous frame, predicting the position and angle of the local coding marker of the next frame and generating a prediction result; as well as The position and angle of the locally encoded marker are corrected by associating the prediction result with the position and angle of the locally encoded marker.

4. The method according to claim 1, wherein: Calculating the final weight of the encoding mark includes: calculating a distance weight based on the position of the coded marker; Calculating a distortion weight based on the distance between each of the corner points and the center point of the encoded mark; and The final weight is calculated by adding the distance weight and the distortion weight.

5. The method according to claim 1, wherein: The coordinates of the corner points are coordinates in a coordinate system in which an origin is a point on the ground perpendicular to each of the plurality of cameras.

6. The method according to claim 1, wherein: Filtering the coded mark includes: determining whether the position of the coded mark exists outside the region of interest; In response to a determination that the location of the coded marker exists outside the region of interest, deleting information about the coded marker; determining whether the width of the encoding mark is different from a preset value; and In response to a determination that the width of the encoding mark is different from the preset value, information about the encoding mark is deleted.

7. The method according to claim 1, wherein: Generating the position and angle of the local encoding marker includes deleting information about encoding markers other than an encoding marker having a highest final weight among encoding markers having the same ID.

8. The method according to claim 2, wherein: Measuring the position of the vehicle includes correcting the position of the vehicle by using the relative distance traveled.

9. A vehicle positioning device, comprising: at least one processor; A memory storing instructions executed by the at least one processor, including the following instructions: identifying corner points of the coded marker from images acquired by each of a plurality of cameras of the vehicle; Calculating the coordinates of the corner points, and calculating the position, width and angle of the coding mark by using the coordinates of the corner points; filtering the coded marks based on the positions and widths of the coded marks; Calculating the final weight of the encoding mark; Converting the angle and position of the coded marker by using the coordinates of one of the plurality of cameras as vehicle coordinates to generate a position and angle of a local coded marker; rotationally transforming the position of the locally encoded marker by using information about the marker in the parking facility and the angle of the locally encoded marker; and The position of the vehicle is measured by using the position of the locally encoded marker and the final weight.

10. The vehicle positioning device according to claim 9, wherein: The instructions executed by the at least one processor include instructions for calculating a relative distance traveled by the vehicle.

11. A non-transitory computer-readable medium having instructions for execution by at least one processor, comprising the following instructions: identifying corner points of the coded marker from images acquired by each of a plurality of cameras of the vehicle; Calculating the coordinates of the corner points, and calculating the position, width and angle of the coding mark by using the coordinates of the corner points; filtering the coded marks based on the positions and widths of the coded marks; Calculating the final weight of the encoding mark; Converting the angle and position of the coded marker by using the coordinates of one of the plurality of cameras as vehicle coordinates to generate a position and angle of a local coded marker; rotationally transforming the position of the locally encoded marker by using information about the marker in the parking facility and the angle of the locally encoded marker; and The position of the vehicle is measured by using the position of the locally encoded marker and the final weight.

Images