Parking garage location tracking method and parking garage location tracking system

Abstract

The invention provides a parking space tracking method and system, and the method comprises the steps: carrying out the tracking of a current parking space detected by a parking space sensor in a current sensing range when an automatic parking system of an automatic driving vehicle is started; judging whether the current parking space exists in a tracking parking space queue of the automatic parking system or not according to the parking space tracking result, the parking space information of the current parking space and the pose information of the automatic driving vehicle in the world coordinate system; if the parking space information exists in the tracking parking space queue, the automatic driving vehicle is controlled to enter parking space selection; the technical problems that in the prior art, when an automatic parking function is used for tracking a parking space, the position of the parking space is usually judged in an image coordinate system, so that the obtained parking space state and position information are inaccurate, a vehicle easily deviates from the parking space during automatic parking, and even collision occurs are solved. And the safety of the automatic driving vehicle in the automatic parking process is influenced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent driving, and more particularly, to a parking space tracking method and a parking space tracking system. Background Art

[0002] Among the rich functions of autonomous driving technology, the automatic parking function is particularly crucial in various parking scenarios. The operation process of the automatic parking function mainly includes the following steps: First is the detection of parking spaces, followed by continuous tracking of the parking spaces, and based on this, the selection of parking spaces. Finally, the automatic parking system will perform autonomous parking operations for the selected parking space. In the parking space tracking link, a unique identification number needs to be assigned to the detected parking spaces in the world coordinate system, and this detected parking space information is data in the image coordinate system extracted by the vision model from the 360-degree panoramic image of the vehicle.

[0003] The parking space tracking and automatic parking of the automatic parking function are dynamic processes that are easily affected by various scenarios. Moreover, the driving speed during the automatic parking process is low, and it is accompanied by a large number of steps such as steering wheel rotation, starting and stopping, and gear shifting. Once there is a slope on the ground or the road condition is poor, it is very easy for the vehicle to deviate when parking into the space or even crash.

[0004] However, there are at least the following problems in the related technologies: When the automatic parking function in the prior art performs parking space tracking, it usually judges the position of the parking space in the image coordinate system, resulting in inaccurate obtained parking space status and position information, and it is easy for the vehicle to deviate from the parking space or even crash during automatic parking, which affects the safety of autonomous driving vehicles during the automatic parking process. Summary of the Invention

[0005] The technical problem solved by the present invention is that when the automatic parking function in the prior art performs parking space tracking, it usually judges the position of the parking space in the image coordinate system, resulting in inaccurate obtained parking space status and position information, and it is easy for the vehicle to deviate from the parking space or even crash during automatic parking, which affects the safety of autonomous driving vehicles during the automatic parking process.

[0006] To solve the above technical problems, the present invention provides a parking space tracking method, which is applied to an autonomous driving vehicle. The parking space tracking method includes: When the autonomous driving vehicle starts the automatic parking system, perform parking space tracking on the current parking space detected by the parking space sensor within the current sensing range; According to the result of the parking space tracking, obtain the parking space information of the current parking space and the pose information of the autonomous driving vehicle in the world coordinate system; Based on the bin location information and the pose information, determine whether the current bin location exists in the tracking bin location queue of the automatic parking system, and the tracking bin location queue includes multiple historical bin location information; If the bin location information exists in the tracking bin location queue, control the autonomous vehicle to enter bin location selection, and control the autonomous vehicle to perform automatic parking according to the result of bin location selection and the pose information.

[0007] Compared with the prior art, the technical effects achieved by adopting this technical solution: This solution tracks the current bin location detected by the bin sensor within the current sensing range when the automatic parking system of the autonomous vehicle is started, and determines whether the current bin location exists in the tracking bin location queue of the automatic parking system according to the bin location information and the pose information of the autonomous vehicle in the world coordinate system, so as to obtain the status and position information of the bin location more accurately, avoid the situation that the vehicle deviates from the bin location or even crashes during automatic parking, thereby improving the safety of the autonomous vehicle during the automatic parking process.

[0008] In an example of the present invention, the bin location information includes the panoramic view output by the vision model and multiple bin corner points of the current bin location. Obtaining the bin location information of the current bin location and the pose information of the autonomous vehicle in the world coordinate system includes: Obtain the picture information of the panoramic view; According to the picture information, obtain the first corner point coordinates of multiple bin corner points in the image coordinate system; According to the pose information and the picture information, convert the first corner point coordinates to the world coordinate system to obtain the second corner point coordinates.

[0009] Compared with the prior art, the technical effects achieved by adopting this technical solution: This solution converts the first corner point coordinates to the world coordinate system to obtain the second corner point coordinates according to the pose information and the picture information, so that the automatic parking system can obtain the accurate position of the current bin location in the world coordinate system, thereby avoiding the situation that the autonomous vehicle deviates from the bin location or even crashes during the automatic parking operation, and further improving the safety of the autonomous vehicle during the automatic parking process.

[0010] In an example of the present invention, the picture information includes the picture length of the panoramic view, the picture width of the panoramic view, and the picture resolution of the panoramic view. According to the picture information, obtaining the first corner point coordinates of multiple bin corner points in the image coordinate system includes: According to the picture length, the picture width, and the picture resolution, calculate the center coordinate value of the image center of the panoramic view in the image coordinate system; Denote the center coordinate value as (Cx, Cy), and the calculation formula for calculating the coordinate value of the image center of the panoramic view in the image coordinate system is: Formula 1: ; Formula 2: ; Wherein, Cx in the formula represents the x - coordinate of the image center in the image coordinate system, Cy represents the y - coordinate of the image center in the image coordinate system, h is the length of the picture, w is the width of the picture, and r is the picture resolution.

[0011] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By calculating the central coordinate value of the image center of the panoramic view in the image coordinate system according to the picture length, picture width, and picture resolution, this solution accurately locates the center point of the image, ensuring the reliability of the automatic parking system in identifying and tracking the parking space, and further improving the safety of the autonomous vehicle during the automatic parking process.

[0012] In an example of the present invention, the pose information includes the current vehicle coordinates and the rotation angle of the autonomous vehicle. According to the pose information and the picture information, converting the first corner point coordinates to the world coordinate system to obtain the second corner point coordinates includes: Obtaining the rotation angle when converting the current vehicle coordinates to the world coordinate system; According to the central coordinate value, calculating the offset of any parking - space corner point relative to the image center when converting the first corner point coordinates to the world coordinate system; Calculating the corner - point coordinate value of any parking - space corner point after being converted to the world coordinate system according to the rotation angle, the offset, and the current vehicle coordinates.

[0013] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By calculating the corner - point coordinate value of any parking - space corner point after being converted to the world coordinate system according to the rotation angle of the vehicle and the offset of the image center, this solution accurately converts the corner - point coordinates in the image coordinate system to the world coordinate system, thereby ensuring that the autonomous vehicle can accurately locate the parking space during automatic parking and improving the accuracy of the parking - space tracking result.

[0014] In an example of the present invention, defining the coordinate value of any parking - space corner point in the image coordinate system as (Px, Py) and defining the offset as B, calculating the offset of the parking - space corner point relative to the image center includes: Calculating the x - axis offset of the parking - space corner point relative to the image center in the x - axis direction, and the calculation formula for the x - axis offset is: Formula 3: ; Calculating the y - axis offset of the parking - space corner point relative to the image center in the y - axis direction, and the calculation formula for the y - axis offset is: Formula 4: ; Wherein, Bx is the x - axis offset and By is the y - axis offset.

[0015] Compared with the prior art, the technical effects achieved by adopting this technical solution: By calculating the x-axis offset of the corner point of the storage location relative to the image center in the x-axis direction and the y-axis offset of the corner point of the storage location relative to the image center in the y-axis direction, this solution can more accurately determine the corner point coordinate values of the storage location in the world coordinate system, thereby improving the accuracy of the storage location tracking result of the autonomous driving vehicle during automatic parking.

[0016] In one example of the present invention, defining the current vehicle coordinates as (Ux, Uy), calculating the corner point coordinate values of any storage location corner point after being transformed into the world coordinate system according to the rotation angle, offset, and current vehicle coordinates includes: Calculating the x-axis coordinate value of any storage location corner point after being transformed into the world coordinate system in the x-axis direction, and the calculation formula for the x-axis coordinate value is: Formula 5: ; Calculating the y-axis coordinate value of any storage location corner point after being transformed into the world coordinate system in the y-axis direction, and the calculation formula for the y-axis coordinate value is: Formula 6: ; Wherein, Wx is the x-axis coordinate value, Wy is the y-axis coordinate value, and a is the rotation angle.

[0017] Compared with the prior art, the technical effects achieved by adopting this technical solution: This solution ensures that the autonomous driving vehicle can accurately perform storage location tracking during parking, reduces the errors during automatic parking, and reduces the risk of collisions and accidents of the autonomous driving vehicle during automatic parking.

[0018] In one example of the present invention, multiple historical storage location information includes multiple historical storage location corner points of multiple historical storage locations in the world coordinate system and multiple historical corner point coordinates of multiple historical storage location corner points. According to the storage location information and pose information, determining whether the current storage location exists in the tracking storage location queue of the automatic parking system includes: Comparing the second corner point coordinates with multiple historical corner point coordinates to determine whether there are historical corner point coordinates in the multiple historical corner point coordinates that match the second corner point coordinates; If so, it is determined that the storage location information exists in the tracking storage location queue, and the current storage location is marked as a matching storage location; If not, it is determined that the storage location information does not exist in the tracking storage location queue, and the current storage location is marked as a new storage location; Wherein, when the storage location information exists in the tracking storage location queue, the second corner point coordinates are added to the tracking storage location queue through a weighted average calculation method to update the multiple historical corner point coordinates; and / or When the storage location information does not exist in the tracking storage location queue, new position information is assigned to the new storage location, and the position information is added to the tracking storage location queue.

[0019] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: This solution reduces the situations of mis-identification and repeated identification, thereby improving the accuracy of the result of bay tracking.

[0020] In an example of the present invention, the parking bay tracking method further includes: Within the current sensing range, determine whether the historical bay is detected according to the matching bay; If not, mark the undetected bay as the target bay, and obtain the current status value of the target bay; According to the current status value, determine whether the current status value satisfies the following formula: Formula 7: ; If it is satisfied, delete the historical bay from the tracking bay queue; If it is not satisfied, there is no need to delete the historical bay from the tracking bay queue: where ω is the current status value.

[0021] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: This solution prevents the accumulation of outdated or invalid data in the tracking bay queue by deleting the undetected bays, ensuring the accuracy of the data in the tracking bay queue.

[0022] On the other hand, the present invention further provides a bay tracking system. The bay tracking system executes the parking bay tracking method in any of the above technical solutions. The bay tracking system includes a bay tracking module, which is used to perform bay tracking on the current bay detected by the bay sensor within the current sensing range when the automatic parking system of the autonomous vehicle is started; a bay information acquisition module, which is used to obtain the bay information of the current bay and the pose information of the autonomous vehicle in the world coordinate system according to the result of the bay tracking; a judgment module, which is used to judge whether the current bay exists in the tracking bay queue of the automatic parking system according to the bay information and the pose information; a control module, which is used to control the autonomous vehicle to enter bay selection when the bay information exists in the tracking bay queue, and control the autonomous vehicle to perform automatic parking according to the result of the bay selection and the pose information.

[0023] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: A bay tracking system provided by this solution executes the parking bay tracking method in any of the above technical solutions. Therefore, this solution has all the beneficial effects described in any of the above technical solutions, which will not be elaborated here.

[0024] In an example of the present invention, the storage location tracking system further includes an image information acquisition module for acquiring the image information of the panoramic view, a coordinate acquisition module for acquiring the first corner coordinates of multiple storage location corner points in the image coordinate system according to the image information, and a coordinate system conversion module for converting the first corner coordinates into the world coordinate system to obtain the second corner coordinates according to the pose information and the image information.

[0025] After adopting the technical solution of the present invention, the following technical effects can be achieved: The present invention performs storage location tracking on the current storage location detected by the storage location sensor within the current perception range when the automatic parking system of the autonomous vehicle is started, and determines whether the current storage location exists in the tracking storage location queue of the automatic parking system according to the storage location information and the pose information of the autonomous vehicle in the world coordinate system, so as to more accurately obtain the status and position information of the storage location, avoid the situation that the vehicle deviates from the storage location or even crashes during automatic parking, thereby improving the safety of the autonomous vehicle during the automatic parking process. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. Figure 1 FIG. 1 is one of the flowcharts of a parking space tracking method provided by an embodiment of the present invention; Figure 2 FIG. 2 is one of the block diagrams of a storage location tracking system provided by an embodiment of the present invention; Figure 3 FIG. 3 is another flowchart of a parking space tracking method provided by an embodiment of the present invention; Figure 4 FIG. 4 is another block diagram of a storage location tracking system provided by an embodiment of the present invention.

[0027] DESCRIPTION OF THE REFERENCE NUMERALS: 100, storage location tracking module; 200, storage location information acquisition module; 300, judgment module; 400, control module; 500, image information acquisition module; 600, coordinate acquisition module; 700, coordinate system conversion module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings.

[0029] As Figure 1As shown in the figure, the present invention provides a method for tracking parking spaces in a parking garage. The method for tracking parking spaces in a parking garage is applied to an autonomous vehicle, and the method for tracking parking spaces includes: S100: When the autonomous vehicle starts the automatic parking system, perform parking space tracking on the current parking space detected by the parking space sensor within the current sensing range; S200: According to the result of the parking space tracking, obtain the parking space information of the current parking space and the pose information of the autonomous vehicle in the world coordinate system; S300: According to the parking space information and the pose information, determine whether the current parking space exists in the tracking parking space queue of the automatic parking system. The tracking parking space queue includes multiple historical parking space information; S400: If the parking space information exists in the tracking parking space queue, control the autonomous vehicle to enter the parking space selection, and control the autonomous vehicle to perform automatic parking according to the result of the parking space selection and the pose information.

[0030] In this solution, when the autonomous vehicle starts the automatic parking system, parking space tracking is performed on the current parking space detected by the parking space sensor within the current sensing range, and according to the parking space information and the pose information of the autonomous vehicle in the world coordinate system, it is determined whether the current parking space exists in the tracking parking space queue of the automatic parking system, so as to more accurately obtain the status and position information of the parking space, avoid the vehicle deviating from the parking space or even crashing during automatic parking, and thus improve the safety of the autonomous vehicle during the automatic parking process.

[0031] As Figure 3 shown, in an example of the present invention, the parking space information includes the panoramic view output by the vision model and multiple parking space corner points of the current parking space. Obtaining the parking space information of the current parking space and the pose information of the autonomous vehicle in the world coordinate system includes: S500: Obtain the picture information of the panoramic view; S600: According to the picture information, obtain the first corner point coordinates of multiple parking space corner points in the image coordinate system; S700: According to the pose information and the picture information, convert the first corner point coordinates to the world coordinate system to obtain the second corner point coordinates.

[0032] In this solution, by converting the first corner point coordinates to the world coordinate system according to the pose information and the picture information to obtain the second corner point coordinates, the automatic parking system can obtain the accurate position of the current parking space in the world coordinate system, thus avoiding the situation that the autonomous vehicle deviates from the parking space or even crashes during the automatic parking operation, and further improving the safety of the autonomous vehicle during the automatic parking process.

[0033] In an example of the present invention, the picture information includes the picture length of the panoramic view, the picture width of the panoramic view, and the picture resolution of the panoramic view. According to the picture information, obtaining the first corner coordinates of multiple storage locations in the image coordinate system includes: Calculating the center coordinate value of the image center of the panoramic view in the image coordinate system according to the picture length, the picture width, and the picture resolution; Denote the center coordinate value as (Cx, Cy). The calculation formula for the coordinate value of the image center of the panoramic view in the image coordinate system is: Formula 1: ; Formula 2: ; Wherein, in the formula, Cx represents the x coordinate of the image center in the image coordinate system, Cy represents the y coordinate of the image center in the image coordinate system, h is the picture length, w is the picture width, and r is the picture resolution.

[0034] This solution calculates the center coordinate value of the image center of the panoramic view in the image coordinate system according to the picture length, the picture width, and the picture resolution to accurately locate the center point of the image, ensuring the reliability of the automatic parking system when identifying and tracking storage locations, and further improving the safety of the autonomous vehicle during the automatic parking process.

[0035] In an example of the present invention, the pose information includes the current vehicle coordinates and the rotation angle of the autonomous vehicle. According to the pose information and the picture information, converting the first corner coordinates to the world coordinate system to obtain the second corner coordinates includes: Obtaining the rotation angle when converting the current vehicle coordinates to the world coordinate system; According to the center coordinate value, calculating the offset of any storage location corner relative to the image center when converting the first corner coordinates to the world coordinate system; Calculating the corner coordinate value of any storage location corner after being converted to the world coordinate system according to the rotation angle, the offset, and the current vehicle coordinates.

[0036] This solution calculates the corner coordinate value of any storage location corner after being converted to the world coordinate system according to the rotation angle of the vehicle and the offset of the image center, so as to accurately convert the corner coordinates in the image coordinate system to the world coordinate system, thereby ensuring that the autonomous vehicle can accurately locate the storage location during automatic parking and improving the accuracy of the storage location tracking result.

[0037] In an example of the present invention, define the coordinate value of any storage location corner in the image coordinate system as (Px, Py), define the offset as B. Calculating the offset of the storage location corner relative to the image center includes: Calculating the x-axis offset of the storage location corner relative to the image center in the x-axis direction, and the calculation formula for the x-axis offset is: Formula 3: ; Calculate the y-axis offset of the corner point of the storage location relative to the image center in the y-axis direction, and the calculation formula for the y-axis offset is: Formula 4: ; where Bx is the x-axis offset and By is the y-axis offset.

[0038] In this solution, by calculating the x-axis offset of the corner point of the storage location relative to the image center in the x-axis direction and the y-axis offset of the corner point of the storage location relative to the image center in the y-axis direction, the corner point coordinate value of the storage location in the world coordinate system can be determined more accurately, thereby improving the accuracy of the storage location tracking result when the autonomous vehicle is parking automatically.

[0039] In an example of the present invention, the current vehicle coordinates are defined as (Ux, Uy). Calculating the corner point coordinate value of any storage location corner point after being transformed into the world coordinate system according to the rotation angle, offset, and current vehicle coordinates includes: Calculate the x-axis coordinate value of any storage location corner point after being transformed into the world coordinate system in the x-axis direction, and the calculation formula for the x-axis coordinate value is: Formula 5: ; Calculate the y-axis coordinate value of any storage location corner point after being transformed into the world coordinate system in the y-axis direction, and the calculation formula for the y-axis coordinate value is: Formula 6: ; where Wx is the x-axis coordinate value, Wy is the y-axis coordinate value, and a is the rotation angle.

[0040] This solution ensures that the autonomous vehicle can accurately track the storage location during parking, reduces the error during the automatic parking process, and reduces the risk of collision and accident of the autonomous vehicle during the automatic parking process.

[0041] In an example of the present invention, multiple historical storage location information includes multiple historical storage location corner points of multiple historical storage locations in the world coordinate system and multiple historical corner point coordinates of multiple historical storage location corner points. Judging whether the current storage location exists in the tracking storage location queue according to the storage location information and the pose information includes: Compare the second corner point coordinates with multiple historical corner point coordinates to judge whether there is a historical corner point coordinate in the multiple historical corner point coordinates that matches the second corner point coordinates; If so, it is determined that the storage location information exists in the tracking storage location queue, and the current storage location is marked as a matching storage location; If not, it is determined that the storage location information does not exist in the tracking storage location queue, and the current storage location is marked as a new storage location; Among them, when the bin location information exists in the tracking bin location queue, the second corner point coordinates are added to the tracking bin location queue through the weighted average calculation method to update multiple historical corner point coordinates; and / or When the bin location information does not exist in the tracking bin location queue, new location information is assigned to the new bin, and the location information is added to the tracking bin location queue. This solution reduces the situations of misidentification and repeated identification, thereby improving the accuracy of the bin location tracking result.

[0042] In an example of the present invention, the parking bin location tracking method further includes: Within the current sensing range, it is determined whether the historical bin location is detected according to the matching bin location; If not, the undetected bin location is marked as the target bin location, and the current state value of the target bin location is obtained; According to the current state value, it is determined whether the current state value satisfies the following formula: Formula 7: ; If it is satisfied, the historical bin location is deleted from the tracking bin location queue; If it is not satisfied, it is not necessary to delete the historical bin location from the tracking bin location queue: Among them, ω is the current state value, and the current state value is usually set to 1 (i.e., ω = 1).

[0043] This solution deletes the undetected bin locations to prevent the accumulation of outdated or invalid data in the tracking bin location queue and ensure the accuracy of the data in the tracking bin location queue.

[0044] On the other hand, as Figure 2 shown, the present invention also provides a bin location tracking system. The bin location tracking system executes the parking bin location tracking method in any of the above technical solutions. The bin location tracking system includes a bin location tracking module 100, a bin location information acquisition module 200, a judgment module 300, and a control module 400. Specifically, the bin location tracking module 100 is used to perform bin location tracking on the current bin detected by the bin location sensor within the current sensing range when the automatic parking system of the autonomous vehicle is started. The bin location information acquisition module 200 is used to obtain the bin location information of the current bin and the pose information of the autonomous vehicle in the world coordinate system according to the result of the bin location tracking. The judgment module 300 is used to judge whether the current bin exists in the tracking bin location queue of the automatic parking system according to the bin location information and the pose information. The control module 400 is used to control the autonomous vehicle to enter bin selection when the bin location information exists in the tracking bin location queue, and control the autonomous vehicle to perform automatic parking according to the result of the bin selection and the pose information.

[0045] A location tracking system provided by this solution, the location tracking system executes the parking location tracking method in any of the above technical solutions. Therefore, this solution has all the beneficial effects described in any of the above technical solutions, which will not be elaborated here.

[0046] As Figure 4 shown, in an example of the present invention, the location tracking system further includes a picture information acquisition module 500, a coordinate acquisition module 600, and a coordinate system conversion module 700. Specifically, the picture information acquisition module 500 is used to acquire the picture information of the panoramic view, the coordinate acquisition module 600 is used to acquire the first corner coordinates of multiple location corner points in the image coordinate system according to the picture information, and the coordinate system conversion module 700 is used to convert the first corner coordinates to the world coordinate system to obtain the second corner coordinates according to the pose information and the picture information.

[0047] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be subject to the scope defined by the claims.

Claims

1. A parking space tracking method, characterized in that, The parking space tracking method is applied to an autonomous vehicle, and the parking space tracking method includes: When the autonomous vehicle starts the automatic parking system, perform parking space tracking on the current parking space detected by the parking space sensor within the current sensing range; According to the result of the parking space tracking, obtain the parking space information of the current parking space and the pose information of the autonomous vehicle in the world coordinate system; According to the parking space information and the pose information, determine whether the current parking space exists in the tracking parking space queue of the automatic parking system, and the tracking parking space queue includes a plurality of historical parking space information; If the parking space information exists in the tracking parking space queue, control the autonomous vehicle to enter parking space selection, and control the autonomous vehicle to perform automatic parking according to the result of the parking space selection and the pose information.

2. The parking space tracking method according to claim 1, wherein The parking space information includes the panoramic view output by the vision model and a plurality of parking space corner points of the current parking space. Obtaining the parking space information of the current parking space and the pose information of the autonomous vehicle in the world coordinate system includes: Obtain the picture information of the panoramic view; According to the picture information, obtain the first corner point coordinates of the plurality of parking space corner points in the image coordinate system; According to the pose information and the picture information, convert the first corner point coordinates to the world coordinate system to obtain the second corner point coordinates.

3. The parking space tracking method according to claim 2, wherein The picture information includes the picture length, the picture width, and the picture resolution of the panoramic view. According to the picture information, obtaining the first corner point coordinates of the plurality of parking space corner points in the image coordinate system includes: According to the picture length, the picture width, and the picture resolution, calculate the center coordinate value of the image center of the panoramic view in the image coordinate system; Denote the center coordinate value as (Cx, Cy), and the calculation formula for the coordinate value of the image center of the panoramic view in the image coordinate system is: Formula 1: ; Formula 2: ; Wherein, in the formula, Cx represents the x coordinate of the image center in the image coordinate system, Cy represents the y coordinate of the image center in the image coordinate system, h is the picture length, w is the picture width, and r is the picture resolution.

4. The parking space tracking method according to claim 3, wherein The pose information includes the current vehicle coordinates and the rotation angle of the autonomous vehicle. According to the pose information and the picture information, converting the first corner point coordinates to the world coordinate system to obtain the second corner point coordinates includes: Obtain the rotation angle when converting the current vehicle coordinates to the world coordinate system; According to the center coordinate value, calculate the offset of any parking space corner point relative to the image center when converting the first corner point coordinates to the world coordinate system; Calculate the corner point coordinate value of any parking space corner point after being converted to the world coordinate system according to the rotation angle, the offset, and the current vehicle coordinates.

5. The parking space tracking method according to claim 4, wherein Define the coordinate value of any parking space corner point in the image coordinate system as (Px, Py), and define the offset as B. Calculating the offset of the parking space corner point relative to the image center includes: Calculate the x-axis offset of the storage location corner point relative to the image center in the x-axis direction, and the calculation formula for the x-axis offset is as follows: Formula 3: ; Calculate the y-axis offset of the storage location corner point relative to the image center in the y-axis direction, and the calculation formula for the y-axis offset is as follows: Formula 4: ; Where, Bx is the x-axis offset, and By is the y-axis offset.

6. The parking space tracking method according to claim 5, wherein, Define the current vehicle coordinates as (Ux, Uy), and calculating the corner point coordinate value of any storage location corner point after being transformed into the world coordinate system according to the rotation angle, the offset, and the current vehicle coordinates includes: Calculate the x-axis coordinate value of any storage location corner point after being transformed into the world coordinate system in the x-axis direction, and the calculation formula for calculating the x-axis coordinate value is as follows: Formula 5: ; Calculate the y-axis coordinate value of any storage location corner point after being transformed into the world coordinate system in the y-axis direction, and the calculation formula for calculating the y-axis coordinate value is as follows: Formula 6: ; Where, Wx is the x-axis coordinate value, Wy is the y-axis coordinate value, and a is the rotation angle.

7. The parking space tracking method according to any one of claims 2 to 6, characterized in that, The multiple historical storage location information includes multiple historical storage location corner points of multiple historical storage locations in the world coordinate system and multiple historical corner point coordinates of the multiple historical storage location corner points. Judging whether the current storage location exists in the tracking storage location queue of the automatic parking system according to the storage location information and the pose information includes: Compare the second corner point coordinates with the multiple historical corner point coordinates, and judge whether there are historical corner point coordinates in the multiple historical corner point coordinates that match the second corner point coordinates; If so, determine that the storage location information exists in the tracking storage location queue, and mark the current storage location as a matching storage location; If not, determine that the storage location information does not exist in the tracking storage location queue, and mark the current storage location as a new storage location; Where, when the storage location information exists in the tracking storage location queue, the second corner point coordinates are added to the tracking storage location queue through a weighted average calculation method to update the multiple historical corner point coordinates; and / or When the storage location information does not exist in the tracking storage location queue, new position information is given to the new storage location, and the position information is added to the tracking storage location queue.

8. The parking space tracking method according to claim 7, wherein The parking space tracking method further includes: Within the current sensing range, judge whether the historical storage location is detected according to the matching storage location; If not, mark the undetected storage location as an object storage location, and obtain the current state value of the object storage location; According to the current state value, judge whether the current state value satisfies the following formula: Formula 7: ; If it is satisfied, delete the historical storage location from the tracking storage location queue; If it is not satisfied, it is not necessary to delete the historical storage location from the tracking storage location queue: Where, ω is the current state value.

9. A location tracking system, characterized in that, The storage location tracking system executes the parking space tracking method according to any one of claims 1-8. The storage location tracking system includes: A storage location tracking module (100), and the storage location tracking module (100) is used to perform storage location tracking on the current storage location detected by the storage location sensor within the current sensing range when the automatic parking system of the autonomous vehicle is started. The bin location information acquisition module (200) is configured to obtain the bin location information of the current bin and the pose information of the autonomous vehicle in the world coordinate system according to the result of the bin location tracking; The judgment module (300) is configured to judge whether the current bin exists in the tracking bin queue of the automatic parking system according to the bin location information and the pose information; The control module (400) is configured to control the autonomous vehicle to enter bin selection when the bin location information exists in the tracking bin queue, and control the autonomous vehicle to perform automatic parking according to the result of the bin selection and the pose information.

10. The bin location tracking system according to claim 9, wherein The bin location tracking system further includes: The picture information acquisition module (500) is configured to acquire the picture information of the surround view; The coordinate acquisition module (600) is configured to acquire the first corner point coordinates of the plurality of bin corner points in the image coordinate system according to the picture information; The coordinate system conversion module (700) is configured to convert the first corner point coordinates to the second corner point coordinates in the world coordinate system according to the pose information and the picture information.

Images