AGV intelligent positioning method and system

Abstract

The invention discloses an AGV intelligent positioning method and system, and relates to the technical field of AGV positioning, and the method comprises the following steps: building ground gray scale identification information in an AGV operation region, obtaining the coordinates of the ground gray scale identification information, and recording the coordinates as identification position information; setting an image acquisition device based on the AGV, and acquiring a ground image of an AGV operation area to obtain first ground image information; performing image multi-threshold processing based on the first ground image information, and performing identification position extraction processing to obtain image position information; performing positioning processing on the AGV according to the image position information to obtain position information of the AGV in the AGV operation area; the method is used for solving the problem that the driving path of the AGV cannot be timely adjusted according to the work task requirement while the positioning precision is ensured and the ground marker is not changed when the AGV is positioned by setting the ground marker in the existing AGV positioning technology.

Description

Technical Field

[0001] The present invention relates to the technical field of AGV positioning, and particularly to an intelligent positioning method and system for an AGV. Background Art

[0002] AGV (Automated Guided Vehicle) positioning technology refers to the technology of determining the precise position and posture of an AGV in a working environment through various technical means and methods; AGV positioning technology is crucial for the autonomous navigation and precise operation of an AGV; it enables the AGV to accurately travel along a predetermined path, precisely dock at a designated location, and complete tasks such as material handling and goods distribution.

[0003] When the existing AGV positioning technology locates the AGV by setting ground markers, the movement range of the AGV is often greatly limited, and once the ground markers are determined, the driving path of the AGV is relatively fixed; the ground markers are usually laid out according to the preset AGV driving path at the initial stage of project implementation; once the positions of these markers are determined, it is equivalent to setting a fixed track for the AGV; during the operation of the AGV, it determines its own position and driving direction by identifying the markers at these fixed positions and can only move forward along the path indicated by the markers; if the path needs to be changed, the layout of the markers must be readjusted; and during the actual operation, some unexpected situations may occur that require temporarily changing the driving path of the AGV; for example, the ground is damaged, there are obstacles blocking, or there are temporary work tasks that require the AGV to go to a specific location; for an AGV relying on ground marker positioning, it is almost impossible to temporarily change the path; because the ground markers are fixed, the AGV can only navigate according to the preset marker information, and even a simple adjustment of a small section of the path requires re-laying the corresponding markers, which is very difficult to complete in a short time, resulting in the AGV being unable to respond to unexpected situations in a timely manner and affecting the work efficiency and the normal progress of the production process; for example, in the patent application with the publication number CN110018633A, a two-dimensional coding design method for AGV positioning and navigation is disclosed, and this solution locates and navigates the AGV by setting markers, which greatly limits the movement range of the AGV and makes the driving path relatively fixed; therefore, when the existing AGV positioning technology locates the AGV by setting ground markers, it is impossible to timely adjust the driving path of the AGV according to the work task requirements while ensuring the positioning accuracy and without changing the ground markers. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems in the prior art to some extent. By establishing ground gray-scale identification information in the AGV operation area and obtaining the coordinates of the ground gray-scale identification information; collecting the ground image of the AGV operation area to obtain the first ground image information; performing image multi-threshold processing and identification position extraction processing to obtain image position information; performing positioning processing on the AGV according to the image position information to obtain the position information of the AGV; so as to solve the problem that when the existing AGV positioning technology locates the AGV by setting ground markers, it is impossible to adjust the driving path of the AGV in a timely manner according to the work task requirements while ensuring the positioning accuracy and without changing the ground markers.

[0005] To achieve the above object, in the first aspect, the present application provides an intelligent positioning method for an AGV, including the following steps:

[0006] Establish ground gray-scale identification information in the AGV operation area and obtain the coordinates of the ground gray-scale identification information, denoted as identification position information;

[0007] Based on the AGV, set an image acquisition device and collect the ground image of the AGV operation area to obtain the first ground image information;

[0008] Based on the first ground image information, perform image multi-threshold processing and identification position extraction processing to obtain image position information;

[0009] Perform positioning processing on the AGV according to the image position information to obtain the position information of the AGV in the AGV operation area.

[0010] Further, establishing ground gray-scale identification information in the AGV operation area and obtaining the coordinates of the ground gray-scale identification information, denoted as identification position information includes the following sub-steps:

[0011] Obtain the operation area of the AGV on the two-dimensional plane, denoted as the AGV operation area, and establish a plane coordinate system in the AGV operation area, denoted as the area coordinate system;

[0012] Set a square with a side length of a1 as the dividing grid, and use the dividing grid to divide the AGV operation area into multiple square grid areas. For any two adjacent square grid areas, extend uniformly from the adjacent boundary to the two corresponding square grid areas; extend a strip-shaped rectangular area with a width of b1, denoted as the identification demarcation area; denote the remaining square area in any one square grid area after removing the identification demarcation area as the identification square area.

[0013] Further, establishing ground gray-scale identification information in the AGV operation area and obtaining the coordinates of the ground gray-scale identification information, denoted as identification position information further includes the following sub-steps:

[0014] Set the gray tolerance as d1, divide the gray range from 0 to 255 into n arithmetic progressions with a tolerance of d1, denoted as h1, h2, ……, hn; Denote h1, h2, ……, h(n - 1) as the identification gray values;

[0015] For any identified square area, denoted as the first identified area, evenly divide the first identified area into four square areas, denoted as small square areas, set the four small square areas to different or the same gray, and ensure that under the set light intensity LX, the gray value of the image collected by the image acquisition device in the corresponding small square area is the identification gray value; For the identification gray values corresponding to the four small square areas, denoted as small square gray values, for any small square gray value, denoted as xh, denote [xh - d2 / 2, xh + d2 / 2] as the small square gray range corresponding to the corresponding small square area; Mark the corresponding four small square gray ranges as the identification gray parameters of the corresponding identified square area;

[0016] Repeat the processing for all identified square areas, and obtain the corresponding identification gray parameters, and ensure that the identification gray parameters of any two identified square areas are not exactly the same;

[0017] Set all identified boundary areas to the same white, and ensure that under the set light intensity LX, the gray value of the identified boundary area in the image collected by the image acquisition device is hn;

[0018] Obtain the geometric centers of all identified square areas, and the position coordinates of the geometric centers of the corresponding small square areas in the area coordinate system, denoted as the identification position information.

[0019] Furthermore, based on the AGV, set up an image acquisition device, and collect the ground image of the AGV operation area, and the first ground image information is obtained through the following sub - steps:

[0020] Denote the vertical projection point of the center of gravity of the AGV in the AGV operation area as the AGV positioning point; Set up a surface light source and two image acquisition devices on the AGV, set the two image acquisition devices to collect the images of two circular areas in front of the AGV respectively, the two circular areas are denoted as the first acquisition area and the second acquisition area respectively, and denote the corresponding centers as the first center and the second center respectively;

[0021] Set the radii of the first acquisition area and the second acquisition area as r1, r1 = [3*a1 / (2*√2)], and set the horizontal distance between the first center and the second center as 3*a1 / 2, and the vertical distance as a1 / 2.

[0022] Further, an image acquisition device is set based on the AGV, and a ground image of the AGV operation area is acquired to obtain the first ground image information, which further includes the following sub-steps:

[0023] The direction in which the AGV moves forward in a straight line is denoted as the vehicle forward direction; the distances from the first center and the second center to the AGV positioning point are set as f1 and f2 respectively; and the angle between the line connecting the first center and the AGV positioning point and the vehicle forward direction is obtained, denoted as the first angle e1; then the angle between the line connecting the second center and the AGV positioning point and the vehicle forward direction is obtained, denoted as the second angle e2; f1, f2, e1, and e2 are denoted as basic acquisition parameters.

[0024] Adjust the irradiation range of the surface light source so that it can completely cover the first acquisition area and the second acquisition area; obtain the ambient light intensity, and set the light intensity of the surface light source according to the ambient light intensity so that the light intensity of the first acquisition area and the second acquisition area is LX; and use two image acquisition devices to simultaneously acquire the grayscale images of the first acquisition area and the second acquisition area, denoted as the first ground image information.

[0025] Further, image multi-threshold processing is performed on the first ground image information, and identification position extraction processing is performed to obtain the image position information, which includes the following sub-steps:

[0026] For all grayscale images of the first ground image information, according to the identification boundary region in the grayscale image, the grayscale image is segmented along the identification boundary region, and the image region corresponding to the complete identification square region in all grayscale images is obtained, denoted as the identification region image; and the position of the center of the circle in all grayscale images in the grayscale image is obtained, denoted as the center image position.

[0027] Any one of the identification gray values h1, h2, ……, h(n - 1) is denoted as hi, and (hi - d1 / 2, hi + d1 / 2] is denoted as the identification gray range of hi.

[0028] For any one identification region image, denoted as the first identification square image, image multi-threshold processing is performed on the first identification square image, including: for the gray value of any one pixel in the first identification image, denoted as g, if h0 ∈ (hi - d1 / 2, hi + d1 / 2], then let g = hi; after completion, the second identification square image is obtained, and the position of the geometric center of the first identification square image in the grayscale image is obtained, denoted as the square center image position.

[0029] Further, image multi-threshold processing is performed on the first ground image information, and identification position extraction processing is performed to obtain the image position information, which further includes the following sub-steps:

[0030] Divide the second - identified square image evenly into four small square regions. For any small square region, denoted as the first small square region, obtain the position of the geometric center of the first small square region in the corresponding grayscale image, denoted as the small - center image position. Repeat to obtain all small - center image positions;

[0031] Then, arrange the grayscale values of all pixel points in the first small square region in ascending order, denoted as the small - square grayscale sequence; Remove the smallest q1% of the grayscale values and the largest q1% of the grayscale values in the small - square grayscale sequence, and calculate the average value of the remaining small - square grayscale sequence, denoted as the small - square recognition parameter of the second - identified square image. Repeat to obtain all small - square recognition parameters, denoted as the square recognition parameter of the second - identified square image; where q1 is a set threshold, and q1 < 40;

[0032] Compare the square recognition parameter with all identification grayscale parameters. Denote the identification grayscale parameter to which all the square recognition parameters can correspond as the membership grayscale parameter, and obtain the identification position information of the identification square region corresponding to the membership grayscale parameter, denoted as the acquisition position information.

[0033] Further, performing image multi - threshold processing on the first ground image information and performing identification position extraction processing to obtain the image position information also includes the following sub - steps:

[0034] Obtain the center - position information according to the acquisition position information, small - center image position, square - center image position, and center - image position, including: Denote any small - center image position or square - center image position as the first square - center position. According to the pixel distance between the first square - center position and the center - image position in the grayscale image and the pixel length of the side length of the identification square region in the corresponding grayscale image, calculate the distance between the first square - center position and the center - image position corresponding in the AGV running area, and according to the acquisition position information of the first square - center position and the relative direction angle with the center - image position in the grayscale image, calculate the distance - position coordinates of the center - image position in the AGV running area, denoted as the center - position coordinates;

[0035] Repeat to obtain the center - position coordinates corresponding to the small - center image position or square - center image position, and calculate the average value, denoted as the center - position information;

[0036] Repeat to obtain the center - position information corresponding to all grayscale images of the first ground image information, denoted as the image position information.

[0037] Further, perform positioning processing on the AGV according to the image position information to obtain the position information of the AGV in the AGV running area, including the following sub - steps:

[0038] Record any one of the center position information in the image position information as the gray center position. Based on the gray center position and the corresponding basic acquisition parameters, calculate the position of the AGV positioning point. Repeat to obtain the positions of the AGV positioning points corresponding to all the center position information and calculate the average value to obtain the position information of the GV in the AGV operation area.

[0039] In a second aspect, the present application provides an intelligent positioning system for an AGV, including a gray scale identification module, an image acquisition module, an image information module, and a position processing module;

[0040] The gray scale identification module is used to establish ground gray scale identification information in the AGV operation area and obtain the coordinates of the ground gray scale identification information, which is recorded as the identification position information;

[0041] The image acquisition module is based on the AGV to set an image acquisition device and acquire the ground image of the AGV operation area to obtain the first ground image information;

[0042] The image information module includes an image processing unit and a position extraction unit. The image processing unit performs image multi-threshold processing based on the first ground image information, and the position extraction unit is used to perform identification position extraction processing to obtain the image position information;

[0043] The position processing module performs positioning processing on the AGV according to the image position information and the identification position information to obtain the position information of the AGV in the AGV operation area.

[0044] The beneficial effects of the present invention: By establishing ground gray scale identification information in the AGV operation area and obtaining the coordinates of the ground gray scale identification information, which is recorded as the identification position information; setting an image acquisition device based on the AGV and acquiring the ground image of the AGV operation area to obtain the first ground image information; performing image multi-threshold processing based on the first ground image information and performing identification position extraction processing to obtain the image position information; performing positioning processing on the AGV according to the image position information to obtain the position information of the AGV in the AGV operation area; it is possible to adjust the driving path of the AGV in a timely manner according to the work task requirements while ensuring the positioning accuracy and without changing the ground markers;

[0045] In the present invention, a square grid is divided in the AGV operation area, and a grayscale identification is set for each square grid. The advantages are that the grayscale identification is easier to be quickly recognized and analyzed, improving the operation efficiency of the AGV, and providing rich path selection for the AGV; the grid path passed through can be dynamically selected according to the actual situation; by collecting images at different regional positions, it is ensured that each collected image contains at least one complete square grid. The advantage is that the complete square grid provides clear and complete identification features, ensuring the positioning accuracy; the small square grayscale range is set as the corresponding identification grayscale parameter. The advantage is that image acquisition may be affected by external factors, resulting in fluctuations in the grayscale values of the collected images. Setting a certain range can improve the positioning fault tolerance and stability without affecting the positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is the principle block diagram of the system of the present invention;

[0047] Figure 2 is the step flowchart of the method of the present invention;

[0048] Figure 3 is the schematic diagram of the identification demarcation area of the present invention;

[0049] Figure 4 is the acquisition schematic diagram of the first acquisition area and the second acquisition area of the present invention;

[0050] Figure 5 is the structural schematic diagram of the electronic device of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0052] Embodiment 1. Please refer to Figure 1 As shown, the present application provides an intelligent positioning system for an AGV, including a grayscale identification module, an image acquisition module, an image information module, and a position processing module;

[0053] The grayscale identification module is used to establish ground grayscale identification information in the AGV operation area and obtain the coordinates of the ground grayscale identification information, denoted as identification position information;

[0054] The grayscale identification module is configured with a grayscale identification strategy, which includes: obtaining the running area of the AGV on a two-dimensional plane, denoted as the AGV running area, and establishing a plane coordinate system in the AGV running area, denoted as the area coordinate system; that is, the running area of the AGV on the ground.

[0055] Please refer to Figure 3 As shown, set a square with side length a1 as the dividing grid, and use the dividing grid to divide the AGV running area into multiple square grid areas. For any two adjacent square grid areas, extend evenly from the adjacent boundary to the two corresponding square grid areas; extend a strip-shaped rectangular area with width b1, denoted as the identification demarcation area; denote the remaining square area in any square grid area after removing the identification demarcation area as the identification square area; the identification demarcation area is used for later image segmentation, and the later image segmentation is performed along the identification demarcation area, so the width b1 does not need to be too large, generally 0.3 to 2 centimeters; for the square grid area at the edge of the AGV running area, even if a certain side is not adjacent to other square grid areas, an identification demarcation area should also be set at the edge.

[0056] Set the grayscale tolerance as d1, divide the grayscale range of 0 - 255 into n arithmetic progressions with a tolerance of d1, expressed as h1, h2, ……, h(n - 1); denote h1, h2, ……, h(n - 1) as the identification grayscale values; that is, h1 = 0, and hn = 255, hn - h(n - 1) may not be equal to d1, but as long as h1, h2, ……, h(n - 1) are ensured; hn is not denoted as an identification grayscale value and is reserved for the identification demarcation area; d1 should not be too small, as being too small will cause the AGV to be unable to distinguish grayscale values that are close in the later stage. Generally, it should be greater than 8.

[0057] For any identified square area, denoted as the first identified area, divide the first identified area evenly into four square areas, denoted as small square areas. Set the four small square areas to different or the same gray levels, and ensure that under a set light intensity LX, the gray value of the image captured by the image acquisition device in the corresponding small square area is the identified gray value. For the identified gray values corresponding to the four small square areas, denoted as small square gray values, for any small square gray value, denoted as xh, denote [xh - d2 / 2, xh + d2 / 2] as the small square gray range corresponding to the small square area. Mark the corresponding four small square gray ranges as the identified gray parameters of the corresponding identified square area, where d2 is the set size for adjusting the gray range, and d2 < d1; generally, d2 is 2 - 6 when it is less than d1; in this embodiment, d2 = 6. For example, if the small square gray value of a certain small square area is 20, then [17, 23] is the small square gray range of this small square area. Image acquisition may be affected by external factors, resulting in fluctuations in the gray values of the captured images. Setting a certain range can improve the positioning fault tolerance and stability without affecting the positioning accuracy.

[0058] Repeat the processing for all identified square areas, and obtain the corresponding identified gray parameters, and ensure that the identified gray parameters of any two identified square areas are not exactly the same. Each identified gray parameter of an identified square area will have four small square gray ranges. When judging whether they are the same, the order relationship is not considered. For example, if an identified gray parameter is [17, 23], [27, 33], [47, 53], and [67, 73], and another identified gray parameter is [27, 33], [17, 23], [67, 73], and [47, 53], then these two identified gray parameters are the same.

[0059] Set all identified demarcation areas to the same white color, and ensure that under a set light intensity LX, the gray value of the identified demarcation area in the image captured by the image acquisition device is hn.

[0060] Obtain the geometric centers of all identified square areas, and the position coordinates of the geometric centers of the corresponding small square areas in the regional coordinate system, denoted as identified position information.

[0061] In the specific implementation process, the setting of the size a1 of the divided grid and the gray tolerance d1 should correspond to the total area of the AGV running area. Under the condition of meeting the area requirement of the AGV running area, d1 should be as large as possible, and a1 should be appropriately small. For example, the total area of the AGV running area is 2000 square meters; if a1 is set to 0.3m * 0.3m, then at least 2223 divided grids are required, and C 17 4 = 2380, so d1 can be set to 13, and h0 = 0, h17 = 221.

[0062] The image acquisition module sets up an image acquisition device based on the AGV and acquires the ground image of the AGV operation area to obtain the first ground image information;

[0063] The image acquisition module is configured with an image acquisition strategy. The image acquisition strategy includes: denoting the vertical projection point of the center of gravity of the AGV in the AGV operation area as the AGV positioning point; that is, regarding the AGV in the AGV operation area as a mass point for positioning and ignoring the shape of the AGV itself; setting up a surface light source and two image acquisition devices on the AGV, and setting the two image acquisition devices to acquire the images of two circular areas in front of the AGV respectively. The two circular areas are denoted as the first acquisition area and the second acquisition area respectively, and the corresponding centers are denoted as the first center and the second center respectively; the two image acquisition devices should be the same as the above image acquisition device, and the settings of the image acquisition device should also be the same;

[0064] Please refer to Figure 4 As shown, set the radii of the first acquisition area and the second acquisition area as r1, r1 = [3*a1 / (2*√2)], and set the horizontal distance between the first center and the second center as 3*a1 / 2 and the vertical distance as a1 / 2; a radius of r1 can ensure that there are at most two complete marked square areas in a circular area to be acquired, and there may be none; while setting two acquisition areas and a specific distance between them can ensure that there is at least one complete marked square area and at most two in the two acquisition areas; ensuring that the acquired image contains at least one complete marked square area has the advantage that the complete marked square area provides clear and complete features, facilitating the accurate extraction of the grid position information by the image recognition algorithm. The complete marked square area can provide clear boundaries and unique gray-scale combination features, distinguishing it from other grids or the environmental background, reducing the ambiguity in the recognition process, and avoiding positioning errors caused by the lack of some grid information;

[0065] Denote the direction of the AGV moving straight forward as the vehicle forward direction; set the distances between the first center and the second center and the AGV positioning point as f1 and f2 respectively; and obtain the angle between the line connecting the first center and the AGV positioning point and the vehicle forward direction, denoted as the first angle e1; then obtain the angle between the line connecting the second center and the AGV positioning point and the vehicle forward direction, denoted as the second angle e2; denote f1, f2, e1, and e2 as the basic acquisition parameters; because the AGV positioning point position is calculated through the center position subsequently;

[0066] Adjust the irradiation range of the surface light source so that it can completely cover the first acquisition area and the second acquisition area; obtain the ambient light intensity, and set the light intensity of the surface light source according to the ambient light intensity so that the light intensities of the first acquisition area and the second acquisition area are LX; and use two image acquisition devices to simultaneously acquire the grayscale images of the first acquisition area and the second acquisition area, which are recorded as the first ground image information; the surface light source can, to a certain extent, ensure that the light irradiated on the first acquisition area and the second acquisition area is parallel light, and the ambient light can also be regarded as parallel light, which is convenient for calculating the light intensity of the surface light source to be set;

[0067] In the specific implementation process, if the size of the acquisition area is increased to ensure that at least the complete identification square area is acquired, it may cause too many complete identification square areas in the image acquired when the AGV travels to a certain position, resulting in a decrease in the resolution of a single identification square area, and the edge information and grayscale information will become blurred, affecting the accuracy of recognition.

[0068] The image information module includes an image processing unit and a position extraction unit. The image processing unit performs image multi-threshold processing based on the first ground image information, and the position extraction unit is used to perform identification position extraction processing to obtain image position information;

[0069] The image processing unit is configured with an image processing strategy, and the image processing strategy includes: for all grayscale images of the first ground image information, according to the identification boundary area in the grayscale image, segment the grayscale image along the identification boundary area, and obtain the image area corresponding to the complete identification square area in all grayscale images, which is recorded as the identification area image; and obtain the position of the center of the circle in the grayscale image of all grayscale images, which is recorded as the center image position;

[0070] Denote any one of the identification grayscale values h1, h2, ……, h(n - 1) as hi, and denote (hi - d1 / 2, hi + d1 / 2] as the identification grayscale range of hi; for example, if d1 = 10 and h2 = 20, then (15, 25] is denoted as the identification grayscale range of h2;

[0071] For any one identification area image, denoted as the first identification square image, perform image multi-threshold processing on the first identification square image, including: for the grayscale value of any one pixel in the first identification image, denoted as g, if g ∈ (hi - d1 / 2, hi + d1 / 2], then let g = hi; after completion, obtain the second identification square image, and obtain the position of the geometric center of the first identification square image in the grayscale image, which is denoted as the square center image position; for example, if g = 19 and g ∈ (15, 25], then let g = 20;

[0072] The second identification square image is evenly divided into four small square regions. For any one of the small square regions, denoted as the first small square region, obtain the position of the geometric center of the first small square region in the corresponding grayscale image, denoted as the small center image position, and repeat to obtain all the small center image positions;

[0073] Then, arrange the grayscale values of all the pixel points in the first small square region in ascending order, denoted as the small square grayscale sequence; remove the smallest q1% of the grayscale values and the largest q1% of the grayscale values in the small square grayscale sequence, and calculate the average value of the remaining small square grayscale sequence, denoted as the small square recognition parameter of the second identification square image. Repeat to obtain all the small square recognition parameters, denoted as the square recognition parameter of the second identification square image; where q1 is the set threshold, q1 < 40; in this embodiment, q1% = 35%; during the image acquisition process, some pixel points with abnormal grayscale values may be generated due to factors such as noise, uneven illumination, and reflection; this will cause great interference to the calculation of the overall grayscale feature; by removing the extreme values at both ends, the remaining grayscale sequence more centrally reflects the grayscale information of the main part;

[0074] The position extraction unit is configured with a position extraction strategy. The position extraction includes: comparing the square recognition parameter with all the identification grayscale parameters, and the identification grayscale parameter to which all the square recognition parameters can all correspond is denoted as the membership grayscale parameter, and obtain the identification position information of the identification square region corresponding to the membership grayscale parameter, denoted as the acquisition position information; all corresponding to belong means that the four square recognition parameters should correspond to the four identification grayscale parameters. Because when setting the identification of the small square region above, the same small square region may exist in one identification square region; for example, a certain square recognition parameter is 20, 19, 30, 50; a certain identification grayscale parameter is [17, 23], [27, 33], [47, 53], and [67, 73]; among them, [67, 73] does not correspond, so this identification grayscale parameter is not the membership grayscale parameter; the order relationship is not considered during the judgment;

[0075] Obtain the center position information according to the acquisition position information, the small center image position, the square center image position, and the center image position, including: denote any one of the small center image positions or the square center image positions as the first square center position, and calculate the distance between the first square center position and the center image position corresponding in the AGV running area according to the pixel distance between the first square center position and the center image position in the grayscale image and the pixel length of the side length of the identification square region in the corresponding grayscale image; for example, a1 - 2*b1 = 30 cm, which is 60 pixel points in the image, that is, each pixel point represents 0.5 cm; the pixel distance between the first square center position and the center image position in the grayscale image is 20 pixel points, that is, the distance is 20 * 0.5 = 10 cm;

[0076] According to the acquisition position information of the first-party center position and the relative direction angle between the center image position and the center of the circle in the grayscale image, the distance position coordinates of the center image position in the AGV operation area are calculated and denoted as the center position coordinates; repeatedly obtain the center position coordinates corresponding to the small center image position or the large center image position, and calculate the average value, which is denoted as the center position information; that is, one set of five center position coordinates is calculated through four small center image positions and one large center image position, and the average value of the five center position coordinates is obtained as the center position information; if there are two complete marked square areas in the collected image and the two complete marked square areas are located within the same circular acquisition area, the average value of the ten center position coordinates obtained through processing needs to be calculated as the center position information.

[0077] In the specific implementation process: in the actual environment, the lighting conditions may be uneven or there may be noise interference, which will cause fluctuations in the grayscale values of the image; the multi-threshold processing of the image can exclude some fluctuating grayscale values caused by environmental factors according to the preset grayscale range, thereby enhancing the adaptability of the recognition process to environmental changes and reducing the impact of environmental factors on the positioning accuracy.

[0078] The position processing module performs positioning processing on the AGV according to the image position information and the marked position information to obtain the position information of the AGV in the AGV operation area.

[0079] The position processing module is configured with a position processing strategy, and the position processing strategy includes: denoting any one of the center position information in the image position information as the grayscale center position, calculating the position of the AGV positioning point according to the grayscale center position and the corresponding basic acquisition parameters, repeatedly obtaining the positions of the AGV positioning points corresponding to all the center position information, and calculating the average value to obtain the position information of the GV in the AGV operation area; in most cases, there is only the coordinate of one center in the center position information, so the final calculation result does not need to be averaged. If there are two complete marked square areas and the two complete marked square areas are located in two circular acquisition areas, there are the coordinates of two centers in the center position information, and the final calculation result needs to be averaged.

[0080] In the specific implementation process, setting different grayscales as identifiers has the following advantages compared to directly using QR codes as identifiers. When a grayscale identifier is worn, it is relatively easy to repair. One only needs to re-spray or patch the worn area. Once a QR code is damaged, it may be necessary to re-produce and replace the QR code label, resulting in relatively high maintenance costs and workload. The grids of grayscale identifiers usually have a relatively large area and relatively simple patterns, so a small amount of stains or dust is not likely to affect their recognition effect. By contrast, QR codes are sensitive to stains and dust. Once partially blocked or contaminated, it may lead to difficult recognition or even inability to recognize. The features of grayscale identifiers are relatively simple, and image recognition algorithms can quickly process and analyze them to achieve rapid determination. In contrast, QR codes contain more information, and the recognition process requires more computing resources and time for decoding and verification. Moreover, grayscale identifiers have strong customizability and are easy to adjust.

[0081] Embodiment 2. Please refer to Figure 2 As shown in the figure, the present application provides an intelligent positioning method for an AGV, including the following steps:

[0082] Step S1: Establish ground grayscale identifier information in the AGV operation area and obtain the coordinates of the ground grayscale identifier information, denoted as identifier position information. Step S1 includes the following sub-steps:

[0083] Step S101: Obtain the operation area of the AGV in the two-dimensional plane, denoted as the AGV operation area, and establish a plane coordinate system in the AGV operation area, denoted as the area coordinate system.

[0084] Step S102: Set a square with a side length of a1 as the dividing grid, and use the dividing grid to divide the AGV operation area into multiple square grid areas. For any two adjacent square grid areas, uniformly extend from the adjacent boundary to the two corresponding square grid areas; extend a strip-shaped rectangular area with a width of b1, denoted as the identifier demarcation area.

[0085] Step S103: Denote the remaining square area in any square grid area except the identifier demarcation area as the identifier square area.

[0086] Step S104: Set a grayscale tolerance of d1, divide the grayscale range of 0 - 255 into n arithmetic progressions with a tolerance of d1, expressed as h1, h2,..., hn; denote h1, h2,..., h(n - 1) as the identifier grayscale values.

[0087] Step S105: For any identified square area, denoted as the first identified area, divide the first identified area evenly into four square areas, denoted as small square areas. Set the four small square areas to different or the same gray levels, and ensure that under the set illumination intensity LX, the gray value of the image collected by the image acquisition device in the corresponding small square area is the identified gray value;

[0088] Step S106: For the identified gray values corresponding to the four small square areas, denoted as small square gray values, for any small square gray value, denoted as xh, denote [xh - d2 / 2, xh + d2 / 2] as the small square gray range corresponding to the corresponding small square area; Mark the corresponding four small square gray ranges as the identified gray parameters of the corresponding identified square area;

[0089] Step S107: Repeat the process for all identified square areas, and obtain the corresponding identified gray parameters, and ensure that the identified gray parameters of any two identified square areas are not exactly the same;

[0090] Step S108: Set all the identified demarcation areas to the same white color, and ensure that under the set illumination intensity LX, the gray value of the identified demarcation area in the image collected by the image acquisition device is hn;

[0091] Step S109: Obtain the geometric centers of all the identified square areas, and the position coordinates of the geometric centers of the corresponding small square areas in the area coordinate system, denoted as the identified position information.

[0092] Step S2: Based on the AGV, set up an image acquisition device, and collect the ground image of the AGV running area to obtain the first ground image information; Step S2 includes the following sub-steps:

[0093] Step S201: Denote the vertical projection point of the center of gravity of the AGV in the AGV running area as the AGV positioning point; Set up a surface light source and two image acquisition devices on the AGV, and set the two image acquisition devices to collect the images of two circular areas in front of the AGV respectively. The two circular areas are denoted as the first acquisition area and the second acquisition area respectively, and denote the corresponding centers as the first center and the second center respectively;

[0094] Step S202: Set the radii of the first acquisition area and the second acquisition area as r1, r1 = [3*a1 / (2*√2)], and set the horizontal distance between the first center and the second center as 3*a1 / 2, and the vertical distance as a1 / 2;

[0095] Step S203, record the direction in which the AGV moves forward in a straight line as the forward direction of the vehicle; set the distances from the first center and the second center to the AGV positioning point as f1 and f2 respectively; and obtain the angle between the line connecting the first center and the AGV positioning point and the forward direction of the vehicle, denoted as the first angle e1; then obtain the angle between the line connecting the second center and the AGV positioning point and the forward direction of the vehicle, denoted as the second angle e2; record f1, f2, e1, and e2 as the basic acquisition parameters;

[0096] Step S204, adjust the illumination range of the surface light source so that it can completely cover the first acquisition area and the second acquisition area; obtain the ambient light intensity, and set the illumination intensity of the surface light source according to the ambient light intensity so that the illumination intensities of the first acquisition area and the second acquisition area are LX; and use two image acquisition devices to simultaneously acquire the grayscale images of the first acquisition area and the second acquisition area, denoted as the first ground image information.

[0097] Step S3, perform image multi-threshold processing on the first ground image information, and perform identification position extraction processing to obtain image position information; Step S3 includes the following sub-steps:

[0098] Step S301, for all grayscale images of the first ground image information, according to the identification boundary area in the grayscale image, segment the grayscale image along the identification boundary area, and obtain the image area corresponding to the complete identification square area in all grayscale images, denoted as the identification area image; and obtain the position of the center of the circle in the grayscale image of all grayscale images, denoted as the center image position;

[0099] Step S302, record any one of the identification gray values h1, h2,..., h(n - 1) as hi, and denote (hi - d1 / 2, hi + d1 / 2] as the identification gray range of hi;

[0100] Step S303, for any one identification area image, denoted as the first identification square image, perform image multi-threshold processing on the first identification square image, including: for the gray value of any one pixel in the first identification image, denoted as g, if h0 ∈ (hi - d1 / 2, hi + d1 / 2], then let g = hi; after completion, obtain the second identification square image, and obtain the position of the geometric center of the first identification square image in the grayscale image, denoted as the square center image position;

[0101] Step S304, evenly divide the second identification square image into four small square areas. For any one small square area, denoted as the first small square area, obtain the position of the geometric center of the first small square area in the corresponding grayscale image, denoted as the small center image position, and repeat to obtain all small center image positions;

[0102] Step S305: Then, arrange the gray values of all the pixel points within the first small square region in ascending order, denoted as the small square gray sequence; remove the smallest q1% of the gray values and the largest q1% of the gray values in the small square gray sequence, and calculate the average value of the remaining small square gray sequence, denoted as the small square recognition parameter of the second identification square image. Repeat to obtain all the small square recognition parameters, denoted as the square recognition parameter of the second identification square image; where q1 is the set threshold, and q1 < 40.

[0103] Step S306: Compare the square recognition parameter with all the identification gray parameters. Denote the identification gray parameter to which all the square recognition parameters can completely correspond as the membership gray parameter, and obtain the identification position information of the identification square region corresponding to the membership gray parameter, denoted as the acquisition position information.

[0104] Step S307: Obtain the center position information based on the acquisition position information, the position of the small center image, the position of the square center image, and the position of the center image, including: Denote any position of the small center image or the square center image as the first square center position. Calculate the distance between the first square center position and the center image position in the AGV operation area according to the pixel distance between the first square center position and the center image position in the gray image and the pixel length of the side length of the identification square region in the corresponding gray image. Then, calculate the distance position coordinates of the center image position in the AGV operation area according to the acquisition position information of the first square center position and the relative direction angle between the first square center position and the center image position in the gray image, denoted as the center position coordinates.

[0105] Step S308: Repeat to obtain the center position coordinates corresponding to the position of the small center image or the square center image, and calculate the average value, denoted as the center position information.

[0106] Step S309: Repeat to obtain the center position information corresponding to all the gray images of the first ground image information, denoted as the image position information.

[0107] Step S4: Perform positioning processing on the AGV according to the image position information to obtain the position information of the AGV in the AGV operation area; Step S4 includes the following sub-steps:

[0108] Step S401: Denote any center position information in the image position information as the gray center position, and calculate the position of the AGV positioning point according to the gray center position and the corresponding basic acquisition parameters.

[0109] Step S402: Repeat to obtain the positions of the AGV positioning points corresponding to all the center position information, and calculate the average value to obtain the position information of the GV in the AGV operation area.

[0110] Example 3, please refer to Figure 5 as shown.Figure 5 The structure schematic diagram of an electronic device is exemplified. The electronic device may include: a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus. The memory stores computer-readable instructions. The processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in an intelligent positioning method of an AGV are run to achieve the following functions: establish ground gray-scale identification information in the AGV running area, and obtain the coordinates of the ground gray-scale identification information, denoted as identification position information; set an image acquisition device based on the AGV, and acquire the ground image of the AGV running area to obtain the first ground image information; perform image multi-threshold processing based on the first ground image information, and perform identification position extraction processing to obtain image position information; perform positioning processing on the AGV according to the image position information to obtain the position information of the AGV in the AGV running area.

[0111] In addition, when the logical instructions in the above-mentioned memory can be implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0112] Embodiment 4. This application also provides a computer-readable storage medium. This application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in an intelligent positioning method of an AGV as described above are run to achieve the following functions: establish ground gray-scale identification information in the AGV running area, and obtain the coordinates of the ground gray-scale identification information, denoted as identification position information; set an image acquisition device based on the AGV, and acquire the ground image of the AGV running area to obtain the first ground image information; perform image multi-threshold processing based on the first ground image information, and perform identification position extraction processing to obtain image position information; perform positioning processing on the AGV according to the image position information to obtain the position information of the AGV in the AGV running area.

[0113] Through the description of the above embodiments, the embodiments of the present invention can be provided as a method, a system or a computer program product. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0114] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The above-described embodiments are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. For another example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be in an electrical, mechanical or other form.

[0115] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An intelligent positioning method for an AGV, characterized in that, It includes the following steps: Establish ground gray-scale identification information in the AGV operation area, and obtain the coordinates of the ground gray-scale identification information, denoted as identification position information; Based on the AGV, set up an image acquisition device, and collect the ground image of the AGV operation area to obtain the first ground image information; Perform image multi-threshold processing on the first ground image information, and perform identification position extraction processing to obtain image position information; Perform positioning processing on the AGV according to the image position information to obtain the position information of the AGV in the AGV operation area.

2. An intelligent positioning method for an AGV according to claim 1, characterized in that, Establish ground gray-scale identification information in the AGV operation area, and obtain the coordinates of the ground gray-scale identification information, denoted as identification position information, including the following sub-steps: Obtain the operation area of the AGV on the two-dimensional plane, denoted as the AGV operation area, and establish a plane coordinate system in the AGV operation area, denoted as the area coordinate system; Set a square with side length a1 as the dividing grid, and use the dividing grid to divide the AGV operation area into multiple square grid areas. For any two adjacent square grid areas, extend evenly from the adjacent boundary to the two corresponding square grid areas; extend a strip-shaped rectangular area with width b1, denoted as the identification demarcation area; denote the remaining square area in any one square grid area after removing the identification demarcation area as the identification square area.

3. An intelligent positioning method for an AGV according to claim 2, characterized in that, Establish ground gray-scale identification information in the AGV operation area, and obtain the coordinates of the ground gray-scale identification information, denoted as identification position information, further including the following sub-steps: Set the gray-scale tolerance as d1, divide the gray-scale range of 0 - 255 into n arithmetic progressions with tolerance d1, expressed as h1, h2, ……, h(n - 1); denote h1, h2, ……, h(n - 1) as the identification gray-scale values; For any one identification square area, denoted as the first identification area, evenly divide the first identification area into four square areas, denoted as small square areas, set the four small square areas to different or the same gray, and ensure that under the set light intensity LX, the gray-scale value of the image collected by the image acquisition device in the corresponding small square area is the identification gray-scale value; for the identification gray-scale values corresponding to the four small square areas, denoted as small square gray-scale values, for any one small square gray-scale value, denoted as xh, denote [xh - d2 / 2, xh + d2 / 2] as the small square gray-scale range corresponding to the small square area; mark the corresponding four small square gray-scale ranges as the identification gray-scale parameters of the corresponding identification square area; where d2 is the set size of the gray-scale range adjustment, and d2 < d1; Repeat the processing for all identification square areas, and obtain the corresponding identification gray-scale parameters, and ensure that the identification gray-scale parameters of any two identification square areas are not exactly the same; Set all identification demarcation areas to the same white, and ensure that under the set light intensity LX, the gray-scale value of the identification demarcation area in the image collected by the image acquisition device is hn; Obtain the geometric centers of all identification square areas, and the position coordinates of the geometric centers of the corresponding small square areas in the area coordinate system, denoted as identification position information.

4. An intelligent positioning method for an AGV according to claim 3, characterized in that, An image acquisition device is set based on the AGV, and the ground image of the AGV operation area is acquired. The first ground image information is obtained through the following sub-steps: The vertical projection point of the center of gravity of the AGV in the AGV operation area is denoted as the AGV positioning point; a surface light source and two image acquisition devices are set on the AGV. The two image acquisition devices are set to acquire the images of two circular areas in front of the AGV respectively. The two circular areas are denoted as the first acquisition area and the second acquisition area respectively, and the corresponding centers are denoted as the first center and the second center respectively; The radii of the first acquisition area and the second acquisition area are set as r1, r1 = [3*a1 / (2*√2)], and the distance between the first center and the second center in the horizontal direction is set as 3*a1 / 2, and the distance in the vertical direction is a1 / 2.

5. An intelligent positioning method for an AGV according to claim 4, characterized in that, An image acquisition device is set based on the AGV, and the ground image of the AGV operation area is acquired. The first ground image information is also obtained through the following sub-steps: The direction in which the AGV moves forward in a straight line is denoted as the trolley forward direction; the distances from the first center and the second center to the AGV positioning point are set as f1 and f2 respectively; and the angle between the line connecting the first center and the AGV positioning point and the trolley forward direction is obtained, denoted as the first angle e1; then the angle between the line connecting the second center and the AGV positioning point and the trolley forward direction is obtained, denoted as the second angle e2; f1, f2, e1 and e2 are denoted as the basic acquisition parameters; Adjust the irradiation range of the surface light source so that it can completely cover the first acquisition area and the second acquisition area; Obtain the ambient light intensity, set the light intensity of the surface light source according to the ambient light intensity, so that the light intensity of the first acquisition area and the second acquisition area is LX; and use the two image acquisition devices to simultaneously acquire the grayscale images of the first acquisition area and the second acquisition area, denoted as the first ground image information.

6. An intelligent positioning method for an AGV according to claim 5, characterized in that, Based on the first ground image information, image multi-threshold processing is performed, and identification position extraction processing is performed to obtain the image position information through the following sub-steps: For all grayscale images of the first ground image information, according to the identification boundary area in the grayscale image, the grayscale image is segmented along the identification boundary area, and the image area corresponding to the complete identification square area in all grayscale images is obtained, denoted as the identification area image; and the positions of the centers of all grayscale images in the grayscale images are obtained, denoted as the center image positions; Any one of the identification gray values h1, h2,..., h(n - 1) is denoted as hi, and (hi - d1 / 2, hi + d1 / 2] is denoted as the identification gray range of hi; For any one identification area image, denoted as the first identification square image, image multi-threshold processing is performed on the first identification square image, including: for the gray value of any one pixel in the first identification image, denoted as g, if g ∈ (hi - d1 / 2, hi + d1 / 2], then let g = hi; After completion, the second identification square image is obtained, and the position of the geometric center of the first identification square image in the grayscale image is obtained, denoted as the square center image position.

7. An intelligent positioning method for an AGV according to claim 6, characterized in that Performing image multi-threshold processing based on the first ground image information and performing identification position extraction processing to obtain the image position information further includes the following sub-steps: Evenly divide the second identification square image into four small square regions. For any one small square region, denoted as the first small square region, obtain the position of the geometric center of the first small square region in the corresponding grayscale image, denoted as the small center image position, and repeat to obtain all small center image positions; Then arrange the grayscale values of all pixel points in the first small square region in ascending order, denoted as the small square grayscale sequence; remove the smallest q1% of the grayscale values and the largest q1% of the grayscale values in the small square grayscale sequence, and calculate the average value of the remaining small square grayscale sequence, denoted as the small square recognition parameter of the second identification square image. Repeat to obtain all small square recognition parameters, denoted as the square recognition parameter of the second identification square image; where q1 is the set threshold, and q1 < 40; Compare the square recognition parameter with all identification grayscale parameters, and denote the identification grayscale parameter to which the square recognition parameter can all correspond as the membership grayscale parameter. Obtain the identification position information of the identification square region corresponding to the membership grayscale parameter, denoted as the acquisition position information.

8. An intelligent positioning method for an AGV according to claim 7, characterized in that, Performing image multi-threshold processing based on the first ground image information and performing identification position extraction processing to obtain the image position information further includes the following sub-steps: Obtain the center position information according to the acquisition position information, the small center image position, the square center image position, and the center image position, including: Denote any one small center image position or square center image position as the first square center position. According to the pixel distance between the first square center position and the center image position in the grayscale image and the pixel length of the side length of the identification square region in the corresponding grayscale image, calculate the distance between the first square center position and the center image position corresponding in the AGV running area; and according to the acquisition position information of the first square center position and the relative direction angle with the center image position in the grayscale image, calculate the distance position coordinates of the center image position in the AGV running area, denoted as the center position coordinates; Repeat to obtain the center position coordinates corresponding to the small center image position or the square center image position, and calculate the average value, denoted as the center position information; Repeat to obtain the center position information corresponding to all grayscale images of the first ground image information, denoted as the image position information.

9. The intelligent positioning method of an AGV according to claim 8, wherein Performing positioning processing on the AGV according to the image position information to obtain the position information of the AGV in the AGV running area further includes the following sub-steps: Denote any one center position information in the image position information as the grayscale center position. According to the grayscale center position and the corresponding basic acquisition parameters, calculate the position of the AGV positioning point. Repeat to obtain the positions of the AGV positioning points corresponding to all center position information, and calculate the average value to obtain the position information of the GV in the AGV running area.

10. An intelligent positioning system for an AGV, which is used to implement an intelligent positioning method for an AGV according to any one of claims 1-9, and is characterized in that, Including a grayscale identification module, an image acquisition module, an image information module, and a position processing module; The grayscale identification module is used to establish ground grayscale identification information in the AGV running area and obtain the coordinates of the ground grayscale identification information, denoted as the identification position information; The image acquisition module is based on the AGV to set up an image acquisition device, and acquires the ground image of the AGV operation area to obtain the first ground image information; The image information module includes an image processing unit and a position extraction unit. The image processing unit performs image multi-threshold processing based on the first ground image information, and the position extraction unit is used to perform identification position extraction processing to obtain image position information; The position processing module performs positioning processing on the AGV according to the image position information and the identification position information to obtain the position information of the AGV in the AGV operation area.

Images