Agv pose adjustment method and system

Abstract

The application provides an AGV pose adjustment method and system. The AGV pose adjustment method comprises the following steps: determining a first transformation matrix according to a first feature point of a parking buffer area; determining a second feature point of the vehicle according to an image captured by a camera, and determining a second transformation matrix according to the second feature point; obtaining a third transformation matrix according to the first transformation matrix and the second transformation matrix; calculating a first pose of the vehicle relative to the parking buffer area according to the third transformation matrix, and adjusting the pose of the AGV according to the first pose. The AGV pose adjustment method and system calculate the first pose of the vehicle in the parking buffer area after the vehicle is parked in the parking buffer area, so that the AGV can adjust its own pose according to the first pose, thereby realizing the coarse alignment of the AGV and the vehicle and completing the parking operation.

Description

TECHNICAL FIELD

[0001] The present disclosure belongs to the field of parking automation equipment, in particular to an AGV pose adjustment method and system. BACKGROUND

[0002] Parking AGV is a new type of mobile robot designed to solve the parking problem in urban development. Because it can effectively save parking space and improve parking efficiency, it is more and more widely used in flat parking lots and stereo garage.

[0003] The existing parking system needs to use an external mechanical platform to actively limit the pose of the vehicle, that is, the user needs to actively adjust the pose of the vehicle and accurately open the vehicle to the specified mechanical platform. This process has certain requirements for the user's driving skills. At the same time, the construction of the mechanical platform also affects the application range of the parking AGV system. This increases the difficulty and risk of parking to some extent, reduces the parking efficiency, limits the parking experience, reduces the flexibility of parking, and narrows the market prospect of the parking AGV system. SUMMARY

[0004] In view of the above, it is necessary to provide an AGV pose adjustment method and system, and the AGV can adjust the pose according to the pose of the vehicle.

[0005] Therefore, the present application first provides an AGV pose adjustment method, comprising:

[0006] determining a first transformation matrix from the parking buffer zone coordinate system to the camera coordinate system according to the first feature point of the parking buffer zone;

[0007] determining a second feature point of the vehicle according to the image of the vehicle in the parking buffer zone shot by the camera, and determining a second transformation matrix from the vehicle coordinate system to the camera coordinate system according to the second feature point;

[0008] obtaining a third transformation matrix from the vehicle coordinate system to the parking buffer zone coordinate system according to the first transformation matrix and the second transformation matrix;

[0009] calculating a first pose of the vehicle relative to the parking buffer zone according to the third transformation matrix, and adjusting the pose of the AGV according to the first pose.

[0010] According to the AGV pose adjustment method, after adjusting the pose of the AGV according to the target pose, the method further comprises:

[0011] when moving the AGV to the parking buffer zone, determining a second pose of the AGV by the positioning module;

[0012] calculating a horizontal deflection angle and a displacement offset of the AGV according to the first pose and the second pose;

[0013] adjusting the pose of the AGV according to at least one of the horizontal deflection angle and the displacement offset.

[0014] According to the AGV pose adjustment method, the first feature point in the parking buffer coordinate system is obtained.

[0015] The first feature point in the parking buffer coordinate system is obtained.

[0016] The first feature point in the image coordinate system is obtained by manual marking.

[0017] The third coordinate of the first feature point in the camera coordinate system is obtained according to the first coordinate and the second coordinate.

[0018] The first transformation matrix is obtained according to the first coordinate and the third coordinate.

[0019] According to the AGV pose adjustment method, the first coordinate of the first feature point in the parking buffer coordinate system is obtained.

[0020] A rectangular space is selected as the parking buffer, and at least one of the four corner points of the parking buffer is determined as the first feature point.

[0021] The origin of the parking buffer coordinate system is set at the center of the parking buffer, and the first coordinate of the first feature point in the parking buffer coordinate system is determined according to the length of the parking buffer along the length direction and the width direction.

[0022] According to the AGV pose adjustment method, the second feature point of the vehicle is determined according to the image of the vehicle in the parking buffer captured by the camera.

[0023] The outline of the vehicle is extracted, and based on the center axis of the outline, a vehicle rectangular frame corresponding to the outline of the vehicle is fitted.

[0024] At least one of the four corner points of the vehicle rectangular frame is determined as the second feature point.

[0025] According to the AGV pose adjustment method, the second transformation matrix from the vehicle coordinate system to the camera coordinate system is determined according to the second feature point.

[0026] obtaining fourth coordinates of the second feature point in the camera coordinate system according to the intrinsic matrix of the camera;

[0027] obtaining actual dimensions of the vehicle rectangular frame according to the coordinates of the second feature point in the camera coordinate system and the height of the camera;

[0028] determining fifth coordinates of the second feature point in the parking buffer coordinate system according to the actual dimensions of the vehicle rectangular frame;

[0029] determining fifth coordinates of the second feature point in the vehicle coordinate system at the current time according to the coordinates of the second feature point in the parking buffer coordinate system, wherein the vehicle coordinate system is arranged in the vehicle, and the origin of the vehicle coordinate system coincides with the origin of the parking buffer coordinate system;

[0030] obtaining a second transformation matrix of the vehicle coordinate system converted to the camera coordinate system at the current time by a PNP method using the fourth coordinates and the fifth coordinates.

[0031] According to the AGV pose adjustment method, the third transformation matrix from the vehicle coordinate system to the parking buffer coordinate system is obtained according to the first transformation matrix and the second transformation matrix, and the third transformation matrix comprises:

[0032] The third transformation matrix is obtained by the following formula

[0033] wherein, the first transformation matrix is, the second transformation matrix is.

[0034] According to the AGV pose adjustment method, the first pose of the vehicle relative to the parking buffer is calculated according to the third transformation matrix, and the first pose comprises:

[0035] wherein, θ r is a horizontal deflection angle of the vehicle in the camera coordinate system, x r and y r are coordinate values of the vehicle in the camera coordinate system.

[0036] According to the AGV pose adjustment method, before the first transformation matrix and the second transformation matrix are determined, the method further comprises:

[0037] dynamically detecting the parking buffer by the camera;

[0038] When it is determined that the parking buffer is in a static state, the first transformation matrix and the second transformation matrix are determined.

[0039] Further, the application also provides an AGV pose adjustment system, comprising:

[0040] A parking buffer for parking a vehicle;

[0041] A camera arranged above the parking buffer for shooting the parking buffer and the vehicle;

[0042] A controller connected to the camera for executing the AGV pose adjustment method.

[0043] Compared with the prior art, the AGV pose adjustment method and system described above, after the vehicle is parked in the parking buffer, determines a first transformation matrix according to a first feature point of the parking buffer, determines a second transformation matrix according to a second feature point of the vehicle shot by the camera, obtains a third transformation matrix through the first transformation matrix and the second transformation matrix, and calculates a first pose of the vehicle in the parking buffer through the third transformation matrix, so that the AGV can adjust its own pose according to the first pose, realizes coarse alignment of the AGV and the vehicle, and completes the parking operation. In this way, in the parking process, the vehicle pose no longer needs to be corrected by an external mechanical platform, which not only reduces the parking difficulty, but also improves the application range of the parking AGV system. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the specific implementation, the following will briefly introduce the drawings needed to be used in the implementation mode description, obviously, the drawings in the following description are some embodiments of the present disclosure, and for those skilled in the art, other drawings can also be obtained according to these drawings without paying creative labor.

[0045] Figure 1 It is a structural schematic diagram of the AGV pose adjustment system.

[0046] Figure 2 It is a flowchart of the parking AGV system.

[0047] Figure 3 It is a flowchart of the AGV pose adjustment method.

[0048] Figure 4 It is a flowchart of determining the first transformation matrix.

[0049] Figure 5 It is a schematic diagram of the parking buffer coordinate system.

[0050] Figure 6 It is a schematic diagram of the camera coordinate system.

[0051] Figure 7 It is a flowchart of determining the second transformation matrix.

[0052] Explanation of main element symbols

[0053] Parking buffer 10 AGV 20 Camera 30 Vehicle 40

[0054] The following detailed description will further describe the present disclosure in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0055] In order to enable a clearer understanding of the above-mentioned purposes, features and advantages of the present disclosure, the present disclosure is described in detail below in conjunction with the drawings and specific embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict. In the following description, a large number of specific details are described in order to facilitate a full understanding of the present disclosure, and the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present disclosure.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present disclosure belongs. The terms used in the specification herein are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure.

[0057] In various embodiments, in order to facilitate the description without limiting the present disclosure, the term "connected" used in the patent application specification and claims of the present disclosure is not limited to physical or mechanical connection, but can include electrical connection, whether direct or indirect. "Up", "down", "below", "left", "right", and the like are only used to represent relative positional relationships, and when the absolute position of the described object changes, the relative positional relationship also changes accordingly.

[0058] Figure 1 is a structural schematic diagram of an AGV pose adjustment system. As shown in Figure 1 , the AGV pose adjustment system includes a parking buffer area, a global vision adjustment system, an AGV local adjustment system, and a controller.

[0059] The parking buffer area is used to park a vehicle. Generally, the area of the parking buffer area is larger than the area of the vehicle, and is used to park the vehicle. After the vehicle is parked in the parking buffer area, the AGV can adjust its attitude and position according to the attitude and position of the vehicle in the parking buffer area, until the pose of the AGV corresponds to the position of the vehicle, so that the AGV can move to the lower side of the vehicle to carry the vehicle, or replace the battery box of the vehicle, and the like.

[0060] The controller includes, but is not limited to, a processor (CPU), a micro control unit (MCU), a server, and the like, which are devices for interpreting computer instructions and processing data in computer software, for controlling the movement, pose adjustment, and the like, of the AGV.

[0061] The global vision adjustment system coarsely estimates the pose of the vehicle after acquiring the global image of the parking buffer area through the camera, and the AGV completes the initial pose adjustment according to the detection result. Specifically, the camera is arranged above the parking buffer area and is used to capture the parking buffer area and the vehicle. By setting the wide angle and height of the camera, the camera captures the entire parking buffer area. The controller can connect the camera and the AGV through wired or wireless means, acquire the image information captured by the camera, and transmit the pose signal of the vehicle calculated by the AGV pose adjustment method to the AGV. The AGV can adjust according to the pose information.

[0062] The AGV local system includes a positioning module arranged on the AGV, which can be a three-dimensional laser radar. In the present embodiment, the positioning module is arranged at the front end of the vehicle body of the AGV, but can also be arranged at other positions as long as it ensures unobstructed detection of the front part. The positioning module scans the vehicle to obtain point cloud data, and the controller processes the collected point cloud data. The AGV further adjusts its pose according to the processing result to correspond to the pose of the vehicle, realizes alignment with the vehicle, and can then move to the lower part of the vehicle to perform the battery replacement operation, or can move to the lower part of the vehicle to hold the vehicle through a clamping rod after obtaining the wheelbase information of the vehicle, and then carry the vehicle to the target parking space to complete the parking operation.

[0063] Figure 2 is a flowchart of the AGV pose adjustment method. As shown in Figure 2 , the AGV pose adjustment method includes the following steps.

[0064] Step S21: After the user sends a parking request, the vehicle is driven into the parking buffer area.

[0065] Step S22: First adjustment of the AGV pose is implemented by the global vision system. Specifically, first, a global image is acquired by the camera. Then, dynamic object detection is performed on the parking buffer area. When it is determined that the parking buffer area is in a static state, a coarse estimation of the vehicle pose is performed to obtain a first pose of the vehicle relative to the parking buffer area, and the pose of the AGV is adjusted according to the first pose. In the process of detecting the first pose of the vehicle and adjusting the pose of the AGV, in order to improve efficiency, the AGV can be moved to the parking buffer area synchronously. However, in other embodiments, the AGV can also be moved after the pose of the AGV is adjusted, and those skilled in the art can set it according to the needs.

[0066] Step S23: Second adjustment of the AGV pose is implemented by the AGV local system. Specifically, when the AGV is moved to the parking buffer area, the second pose of the AGV is determined by the positioning module. Then, the horizontal deflection angle and the displacement offset of the AGV are calculated according to the first pose and the second pose. Next, the pose of the AGV is adjusted according to at least one of the horizontal deflection angle and the displacement offset, so that the AGV is more accurately positioned relative to the vehicle. Then, the AGV moves to the lower side of the vehicle after acquiring the wheelbase of the vehicle by the positioning module.

[0067] Step S24: The AGV moves to the lower side of the vehicle to complete the related operation. In some embodiments, the AGV clamps the tire of the vehicle after moving to the lower side of the vehicle, and the vehicle is carried to complete the parking operation; in other embodiments, the AGV completes the operation of storing or taking the battery box by moving to the lower side of the vehicle.

[0068] In this embodiment, in step S22, the AGV pose adjustment method is used to acquire the first pose of the vehicle relative to the parking buffer area by the camera, and the first adjustment (coarse adjustment) of the pose of the AGV is completed. The following will be described in combination with Figure 3 the AGV pose adjustment method.

[0069] Figure 3 is a flowchart of the AGV pose adjustment method. As Figure 3 shown, the AGV pose adjustment method includes steps S31-S35.

[0070] Step S31: A first transformation matrix is determined from the parking buffer area coordinate system to the camera coordinate system according to a first feature point of the parking buffer area. Specifically, Figure 4 is a flowchart for determining the first transformation matrix, and this step includes steps S311-S315, which will be described in combination with Figure 4 .

[0071] Step S311: The first coordinate of the first feature point in the parking buffer area coordinate system is acquired. Figure 5is a schematic diagram of a parking buffer coordinate system, as shown in Figure 5 The parking buffer coordinate system is arranged in the parking buffer. A rectangular space is selected to determine the parking buffer, and at least one of the four corner points of the parking buffer is determined as the first feature point. In this embodiment, the four corner points of the parking buffer are selected as the first feature points.

[0072] In step S312, the origin P b of the parking buffer coordinate system is arranged at the center of the parking buffer, and the first coordinates of the first feature points (i.e., the four corner points) are determined according to the length W in the length direction and the length H in the width direction of the parking buffer. Specifically, in the parking buffer coordinate system, the first coordinates of the four corner points are:

[0073] Therefore, formula 1 can be obtained: wherein, In the formula, a k is the weight coefficient of the kth corner point.

[0074] In step S313, the second coordinates of the first feature points in the image coordinate system are obtained by manual marking, wherein the image coordinate system is arranged in the image captured by the camera.

[0075] Specifically, Figure 6 is a schematic diagram of a camera coordinate system solution principle, as shown in Figure 6 In the image coordinate system, the same as the parking buffer coordinate system, formula 2 exists: wherein,

[0076] Through the manual marking method, the four second coordinates of the first feature points in the image coordinate system can be obtained as:

[0077] In step S314, the third coordinates of the first feature points in the camera coordinate system are obtained according to the first coordinates and the second coordinates. Specifically, please continue to refer to Figure 5 According to the pinhole camera model, formula 3 can be obtained:

[0078]

[0079] wherein, f u , f v , u, and v are all internal parameters of the camera; is the coordinate value of the point in the camera coordinate system.

[0080] Formula 3 is written as matrix M 2×12 ·X12×1 The formula 4 is as follows:

[0081]

[0082] The matrix M is decomposed by SVD to obtain the formula 5: 2×12

[0083] M = UΣΣ T , M T M = V(Σ T Σ)V T ;

[0084] Therefore, the X-axis coordinate of the first feature point can be expressed as the formula 6:

[0085]

[0086] Since the distance between any two points in space is equal in different coordinate systems, the formula 7 can be obtained:

[0087]

[0088] The parameter β k is constructed to build an error function formula 8:

[0089]

[0090] From the first-order Taylor expansion, the formula 9: E(β0+Δβ) = E(β0)+E'(β0)Δβ = 0 and the formula 10: E'(β0)Δβ = -E(β) = ρ-Lβ0 can be obtained.

[0091] The error of the formula 8 is minimized by the Gauss-Newton method to obtain the optimal solution β k . Substituting it into the formula 6 can obtain X to obtain the X-axis coordinate of the first feature point. Repeating the above process can obtain the Y-axis coordinate of the first feature point, so that the third coordinate of the first feature point (i.e. the four corner points) of the parking buffer in the camera coordinate system can be obtained.

[0092] Step S315: obtaining the first transformation matrix according to the first coordinate and the third coordinate. In this step, since the first coordinate of the first feature point of the parking buffer in the parking buffer coordinate system is prior, i.e. is known, the coordinates of the four corner points in the parking buffer coordinate system and the camera coordinate system can be obtained respectively (formula 11):

[0093]

[0094] Then, the coordinates of the first feature point in the parking buffer coordinate system relative to the center point are calculated (formula 12):​

[0095]

[0096] Next, the matrix H (formula 13) is calculated:

[0097] Finally, singular value decomposition (SVD) of the matrix H can obtain the rotation matrix R and the translation vector t between the parking buffer coordinate system and the camera coordinate system (formula 14):

[0098] H = UΣΣ, R = VU T , t = P c -RP b , i.e., the transformation matrix from the parking buffer coordinate system to the camera coordinate system is obtained

[0099] Step S32: determining a second feature point of the vehicle according to an image of the vehicle in the parking buffer area captured by the camera, and determining a second transformation matrix from the vehicle coordinate system to the camera coordinate system according to the second feature point.

[0100] In this step, the contour of the vehicle is extracted, and a vehicle rectangular frame corresponding to the contour of the vehicle is fitted based on the center axis of the contour. Then, at least one of the four corner points of the vehicle rectangular frame is determined as the second feature point. Specifically, Figure 7 is a flowchart for determining the second transformation matrix, as Figure 7 determining a second transformation matrix from the vehicle coordinate system to the camera coordinate system according to the second feature point includes steps S321-S324.

[0101] Step S321: obtaining fourth coordinates of the second feature point in the camera coordinate system according to the intrinsic matrix of the camera. Specifically, after the vehicle rectangular frame is extracted, the coordinates of the four corner points (i.e., the second feature points) in the image coordinate system can be obtained. According to the intrinsic matrix of the camera, the fourth coordinates of the second feature points in the camera coordinate system can be obtained.

[0102] Step S322: obtaining the actual size of the vehicle rectangular frame as MxN according to the coordinates of the second feature points in the camera coordinate system and the height of the camera.

[0103] Step S323: determining the coordinates of the second feature points in the parking buffer coordinate system according to the actual size of the vehicle rectangular frame, and determining the fifth coordinates of the second feature points in the vehicle coordinate system at the current time according to the coordinates of the second feature points in the parking buffer coordinate system, wherein the vehicle coordinate system is arranged in the vehicle, and the origin of the vehicle coordinate system coincides with the origin of the parking buffer coordinate system. Specifically, assuming that the actual size of the extracted vehicle rectangular frame ABCD is MxN, the prior coordinates of the four second feature points of the vehicle rectangular frame in the parking buffer coordinate system are Ab(-0.5M, 0.5N), Bb(-0.5M, -0.5N), Cb(0.5M, -0.5N), and Db(0.5M, 0.5N). At the current time, the four corner points of the vehicle rectangular frame correspond to the four points A, B, C, and D in the image coordinate system after mapping, so that the fifth coordinates of the second feature points in the vehicle coordinate system at the current time are obtained.

[0104] Step S324: using the fourth coordinates and the fifth coordinates, and obtaining the second transformation matrix of the conversion from the vehicle coordinate system to the camera coordinate system at the current time by the PNP method.

[0105] Step S33: obtaining the third transformation matrix of the conversion from the vehicle coordinate system to the parking buffer coordinate system according to the first transformation matrix and the second transformation matrix.

[0106] In this step, in the parking AGV global system, the camera is fixed and immovable, so that the third transformation matrix at the current position can be obtained by the following formula.

[0107] wherein, the first transformation matrix is, the second transformation matrix is.

[0108] Step S34: calculating the first pose of the vehicle relative to the parking buffer according to the third transformation matrix. In this step, the first pose P can be calculated by the following formula. r :

[0109]

[0110] wherein, θ r is the horizontal deflection angle of the vehicle in the camera coordinate system, x r and y r are the coordinate values of the vehicle in the camera coordinate system.

[0111] Step S35: adjusting the pose of the AGV according to the first pose, and completing the first pose adjustment of the AGV.

[0112] In this embodiment, in step S23, the AGV local system implements AGV pose accurate adjustment to determine the second pose of the AGV. The AGV second pose adjustment method is described in detail below, and the AGV pose adjustment method includes steps S41-S45.

[0113] Step S41: When the parking AGV reaches the parking buffer area, it is assumed that the center of the parking buffer area is on the left-right middle vertical plane of the AGV, wherein the current pose of the AGV can be obtained in real time by the positioning module thereon, and the current pose of the AGV obtained by the positioning module is set as P = [θx y] a a a a T , and the first pose of the AGV when it reaches the parking buffer area is P0 = [θ0x0 y0] T .

[0114] Step S42: The allowed error of the horizontal deflection angle during adjustment is e1, and the allowed error of the left-right offset is e2. When the horizontal deflection angle and the left-right offset of the AGV relative to the vehicle are both within the allowed error range, the initial pose adjustment of the parking AGV is completed.

[0115] The AGV pose adjustment method and system described above can determine a first transformation matrix according to the first feature point of the parking buffer area after the vehicle is parked in the parking buffer area, determine a second transformation matrix according to the second feature point of the vehicle captured by the camera, obtain a third transformation matrix through the first transformation matrix and the second transformation matrix, and calculate the first pose of the vehicle in the parking buffer area through the third transformation matrix, so that the AGV can adjust its own pose according to the first pose, realize the coarse alignment of the AGV and the vehicle, and complete the parking operation. In this way, in the parking process, it is no longer necessary to correct the vehicle pose through an external mechanical platform, and the vehicle can be parked in the parking buffer area, and the area of the parking buffer area is not limited, which can not only reduce the use requirements of users and optimize the user experience, but also increase the application range of the parking AGV system.

[0116] ​​​​In several embodiments provided by the present disclosure, it is obvious to those skilled in the art that the present disclosure is not limited to the details of the above exemplary embodiments, and the present disclosure can be implemented in other specific forms without departing from the spirit or essential characteristics of the present disclosure. Therefore, the embodiments should be considered as exemplary and non-limiting, and the scope of the present disclosure is defined by the appended claims rather than the above description, and therefore all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present disclosure. In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The words "first", "second" and the like are used to indicate names, not any particular order.

[0117] The above embodiments are only used to illustrate the technical solutions of the present disclosure rather than limit the present disclosure. Although the present disclosure is described in detail with reference to the above preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present disclosure can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present disclosure.

Claims

1. An AGV pose adjustment method, characterized in that, The method comprises: determining a first transformation matrix from the parking buffer coordinate system to the camera coordinate system according to a first feature point of the parking buffer; determining a second feature point of the vehicle according to an image of the vehicle in the parking buffer captured by the camera, and determining a second transformation matrix from the vehicle coordinate system to the camera coordinate system according to the second feature point; obtaining a third transformation matrix from the vehicle coordinate system to the parking buffer coordinate system according to the first transformation matrix and the second transformation matrix; calculating a first pose of the vehicle relative to the parking buffer according to the third transformation matrix, and adjusting the pose of the AGV according to the first pose; determining a second feature point of the vehicle according to an image of the vehicle in the parking buffer captured by the camera comprises: extracting the contour of the vehicle, fitting a vehicle rectangular frame corresponding to the contour of the vehicle based on the center axis of the contour; determining at least one of the four corner points of the vehicle rectangular frame as the second feature point; determining a second transformation matrix from the vehicle coordinate system to the camera coordinate system according to the second feature point comprises: obtaining fourth coordinates of the second feature point in the camera coordinate system according to the intrinsic matrix of the camera; obtaining the actual size of the vehicle rectangular frame according to the coordinates of the second feature point in the camera coordinate system and the height of the camera; determining fifth coordinates of the second feature point in the parking buffer coordinate system according to the actual size of the vehicle rectangular frame; determining the fifth coordinates of the second feature point in the vehicle coordinate system at the current time according to the coordinates of the second feature point in the parking buffer coordinate system, wherein the vehicle coordinate system is arranged in the vehicle, and the origin of the vehicle coordinate system coincides with the origin of the parking buffer coordinate system; obtaining the second transformation matrix from the vehicle coordinate system to the camera coordinate system at the current time by PNP method using the fourth coordinates and the fifth coordinates.

2. The AGV pose adjustment method of claim 1, wherein, After adjusting the pose of the AGV according to the first pose, the method further comprises: moving the AGV to the parking buffer, determining a second pose of the AGV by the positioning module; calculating the horizontal deflection angle and the displacement offset of the AGV according to the first pose and the second pose; adjusting the pose of the AGV according to at least one of the horizontal deflection angle and the displacement offset.

3. The AGV pose adjustment method of claim 1, wherein, Determining a first transformation matrix from the parking buffer coordinate system to the camera coordinate system according to a first feature point of the parking buffer comprises: obtaining first coordinates of the first feature point in the parking buffer coordinate system, wherein the parking buffer coordinate system is arranged in the parking buffer; obtaining second coordinates of the first feature point in the image coordinate system by manual marking, wherein the image coordinate system is arranged in the image captured by the camera; obtaining third coordinates of the first feature point in the camera coordinate system according to the first coordinates and the second coordinates; obtaining the first transformation matrix according to the first coordinates and the third coordinates.

4. The AGV pose adjustment method of claim 3, wherein, Obtaining first coordinates of the first feature point in the parking buffer coordinate system comprises: Selecting a rectangular space as the parking buffer zone, and determining at least one of the four corner points of the parking buffer zone as the first feature point; Setting the origin of the parking buffer zone coordinate system at the center of the parking buffer zone, and determining the first coordinate of the first feature point in the parking buffer zone coordinate system according to the length of the parking buffer zone along the length direction and the width direction.

5. The AGV pose adjustment method of claim 1, wherein, According to the first transformation matrix and the second transformation matrix, a third transformation matrix from the vehicle coordinate system to the parking buffer zone coordinate system is obtained, including: The third transformation matrix is obtained by the following formula wherein, is a first transformation matrix, is a second transformation matrix.

6. The AGV pose adjustment method of claim 5, wherein, According to the third transformation matrix, a first pose of the vehicle relative to the parking buffer zone is calculated, including: where θ r is the horizontal deflection angle of the vehicle in the camera coordinate system, x r and y r are coordinate values of the vehicle in the camera coordinate system.

7. The AGV pose adjustment method of claim 1, wherein, Before determining the first transformation matrix and the second transformation matrix, further including: Performing dynamic detection on the parking buffer zone by the camera; When it is determined that the parking buffer zone is in a static state, the first transformation matrix and the second transformation matrix are determined.

8. An AGV pose adjustment system, characterized in that, Including: A parking buffer zone for parking a vehicle; A camera arranged above the parking buffer zone for capturing the parking buffer zone and the vehicle; A controller connected to the camera for executing the AGV pose adjustment method according to any one of claims 1-7.

Images