Visual guidance methods, devices, electronic equipment and storage media for guided transport vehicles

Abstract

This invention relates to the field of visual guidance technology for transport vehicle routes, and more particularly to a method, device, electronic device, and storage medium for guiding transport vehicles using visual guidance. The method first acquires a target image, a first waypoint, and a target travel point; then, it performs differential processing on the target image and feeds the processed data into an artificial neural network model for recognition, determining the positions of the first waypoint, the target travel point, and the guide line in the target image; next, it determines the current position based on the positions of the first waypoint, the target travel point, and the guide line in the target image; finally, it adjusts the travel direction based on the current position. This invention allows transport vehicles to deviate from the guide line to a certain extent, thereby avoiding oncoming transport vehicles without requiring some transport vehicles to wait for avoidance, improving overall transport efficiency and reducing the number of guide lines.

Description

Technical Field

[0001] This invention relates to the field of visual guidance technology for transport vehicle routes, and in particular to a visual guidance method, device, electronic device, and storage medium for guiding transport vehicles. Background Technology

[0002] AGV is an abbreviation for Automated Guided Vehicle. It refers to a transport vehicle equipped with electromagnetic or optical automatic guidance devices, capable of following a prescribed guidance path, and possessing safety protection and various transfer functions. AGV is a special application of wheeled mobile robots.

[0003] In terms of AGV path guidance methods, there are mainly electromagnetic guidance, inertial guidance, laser guidance, and vision guidance. Among these, vision guidance is favored because it has the advantages of low hardware investment, wide applicability, and minimal modification to the application scenario compared to other guidance methods. A common vision guidance method is to set guide lines on the travel path and place multiple QR codes on the guide lines.

[0004] This approach has low requirements for visual algorithms; however, when multiple AGVs operate on a fixed route, route planning and vehicle scheduling issues need to be considered, which limits the carrying capacity of multiple AGVs. On the other hand, AGVs that use image recognition algorithms to identify the surrounding environment and automatically avoid obstacles require more effort in terms of computing power and the construction of recognition models.

[0005] Therefore, it is necessary to develop and design a visual guidance method for guided transport vehicles. Summary of the Invention

[0006] The present invention provides a visual guidance method, device, electronic device and storage medium for guided transport vehicles, which solves the problem of transport vehicles traveling by guide lines affecting transport capacity in the prior art.

[0007] In a first aspect, embodiments of the present invention provide a visual guidance method for a guided transport vehicle, comprising:

[0008] Acquire the target image, the first waypoint, and the target point;

[0009] The target image is subjected to differential processing, and the processed data is fed into an artificial neural network model for recognition to determine the positions of the first waypoint, the target point, and the guide line in the target image.

[0010] The current position is determined based on the position of the first waypoint, the target point, and the guide line in the target image;

[0011] Adjust the direction of travel based on the current position.

[0012] In one possible implementation, the step of performing differential processing on the target image and feeding the processed data into an artificial neural network model for recognition to determine the positions of the first waypoint, the target point, and the guide line in the target image includes:

[0013] The target image is desaturated to obtain a first grayscale image;

[0014] The distribution range of multiple pixel values ​​in the first grayscale image is adjusted to obtain the second grayscale image;

[0015] Perform differential and pooling operations on the second grayscale image to obtain the first data matrix;

[0016] The first data matrix is ​​fed into the artificial neural network in regions to identify the first waypoint, the target point, and the guide line;

[0017] Based on the recognition results, the positions of the first waypoint, the target point, and the guide line in the target image are extracted.

[0018] In one possible implementation, adjusting the distribution range of multiple pixel values ​​in the first grayscale image to obtain the second grayscale image includes:

[0019] Multiple standard pixel value intervals are obtained, wherein each standard pixel value interval has a predetermined number of continuously distributed pixels;

[0020] Arrange multiple pixel values ​​in the first grayscale image in ascending or descending order to obtain a pixel value queue;

[0021] The pixel value queue is divided according to a predetermined number to obtain multiple first pixel value intervals;

[0022] Based on the plurality of standard pixel value ranges, the plurality of first pixel value ranges, and the first formula, the first grayscale image is adjusted to obtain a second grayscale image, wherein the first formula is:

[0023]

[0024] In the formula, For the nth pixel in the first pixel value range, It is the smallest pixel value in the nth interval of the first pixel value. SP is the largest pixel value in the nth interval of the first pixel value. nMax SP is the largest pixel value in the nth standard pixel value range. nMinIt is the smallest pixel value in the nth standard pixel value range. This refers to the pixel in the adjusted nth first pixel value range.

[0025] In one possible implementation, the step of performing differential and pooling operations on the second grayscale image to obtain the first data matrix includes:

[0026] Obtain the difference operator data block;

[0027] Data blocks of the same type as the difference operator data blocks are extracted multiple times from the second grayscale image according to a predetermined positional order, and used as multiple intermediate data blocks;

[0028] Calculate the dot product between the difference operator data block and the plurality of intermediate data blocks respectively to obtain multiple difference results;

[0029] Based on the position of the intermediate data block corresponding to the difference result in the second grayscale image, the multiple difference results are constructed into an intermediate data matrix;

[0030] Max pooling is performed on the intermediate data matrix to obtain the first data matrix.

[0031] In one possible implementation, the training process of the artificial neural network includes:

[0032] Multiple data blocks are obtained, which are obtained by performing differential and pooling operations on the grayscale image. Each data block corresponds to a waypoint label or guide line label.

[0033] The multiple data blocks are respectively input into the basic model of the artificial neural network to obtain multiple recognition outputs;

[0034] Analyze the deviation between the multiple identification outputs and the labels of the multiple data blocks;

[0035] If the deviation is greater than the recognition error threshold, then the backpropagation algorithm is used to adjust multiple parameters of the artificial neural network base model according to the deviation;

[0036] Otherwise, fix multiple parameters of the artificial neural network base model and use the artificial neural network base model as the artificial neural network.

[0037] In one possible implementation, determining the current position based on the first waypoint, the target point, and the position of the guide line in the target image includes:

[0038] A positioning model is obtained, wherein the positioning model outputs an indication of the current position based on the first waypoint, the target point, and the position of the guide line in the target image;

[0039] The positions of the first waypoint, the target point, and the guide line in the target image are input into the positioning model to obtain the current position;

[0040] The positioning model is determined based on multiple positioning datasets. Each positioning dataset includes current location sample data, first waypoint sample data, target point sample data, and position data of guide line sample data in the sample image.

[0041] The multiple location datasets are input into the location model respectively to obtain multiple location outputs;

[0042] Based on the multiple positioning outputs and the current location sample data of the multiple positioning datasets, the positioning deviation of the positioning model is determined;

[0043] If the positioning deviation is greater than the positioning deviation threshold, then multiple parameters of the positioning model are adjusted according to the positioning deviation.

[0044] In one possible implementation, the positioning model is:

[0045]

[0046] In the formula, P(x) is the x-axis coordinate of the current position, mmax is the total exponent, and w xm c is the weight of the m-th x-axis. x Let P(y) be the x-axis offset constant, and P(y) be the y-axis coordinate indicating the current position. ym c is the weight of the m-th y-axis. y Let w be the y-axis bias constant, X be the joint variable, and w be the y-axis bias constant. b Let x be the weight of the b-th sample data. b This represents the position of the b-th sample image.

[0047] Secondly, embodiments of the present invention provide a visual guidance device for a guided transport vehicle, used to implement the visual guidance method for a guided transport vehicle as described in the first aspect or any possible implementation thereof, the visual guidance device for the guided transport vehicle comprising:

[0048] The data acquisition module is used to acquire the target image, the first waypoint, and the target point.

[0049] The image recognition module is used to perform differential processing on the target image and send the processed data into an artificial neural network model for recognition to determine the positions of the first waypoint, the target point, and the guide line in the target image.

[0050] The positioning module is used to determine the current position based on the position of the first waypoint, the target point, and the guide line in the target image;

[0051] as well as,

[0052] The navigation module is used to adjust the direction of travel based on the current location.

[0053] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor executes the computer program to implement the steps of the method as described in the first aspect or any possible implementation of the first aspect.

[0054] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method as described in the first aspect or any possible implementation thereof.

[0055] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows:

[0056] This invention discloses a visual guidance method for a guided transport vehicle. First, it acquires a target image, a first waypoint, and a target travel point. Then, it performs differential processing on the target image and feeds the processed data into an artificial neural network model for recognition, determining the positions of the first waypoint, the target travel point, and the guide line within the target image. Next, based on the positions of these elements, the current position is determined. Finally, the travel direction is adjusted according to the current position. This invention, while primarily relying on guide lines for guidance, determines the current position through image recognition. Therefore, it allows the transport vehicle to deviate from the guide lines to a certain extent, thus avoiding oncoming transport vehicles without requiring some vehicles to wait for avoidance, improving overall transport efficiency and reducing the number of guide lines. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a flowchart of the visual guidance method for a guided transport vehicle provided in an embodiment of the present invention;

[0059] Figure 2 This is a schematic diagram illustrating the principle of grayscale image pixel value adjustment provided in the embodiments of the present invention;

[0060] Figure 3 This is a schematic diagram of the second grayscale image differential processing process provided in the embodiments of the present invention;

[0061] Figure 4 This is a functional block diagram of the visual guidance device for the guided transport vehicle provided in an embodiment of the present invention;

[0062] Figure 5 This is a functional block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0063] In the following description, specific details such as particular system structures and techniques are set forth for illustrative purposes and not for limitation, so as to provide a thorough understanding of embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0064] To make the objectives, technical solutions, and advantages of the present invention clearer, specific embodiments will be described below in conjunction with the accompanying drawings.

[0065] The embodiments of the present invention will be described in detail below. This example is implemented based on the technical solution of the present invention, and provides detailed implementation methods and specific operation processes. However, the protection scope of the present invention is not limited to the following embodiments.

[0066] Figure 1 A flowchart of a visual guidance method for a guided transport vehicle provided in an embodiment of the present invention.

[0067] like Figure 1 The diagram illustrates the implementation flowchart of the visual guidance method for a guided transport vehicle provided by an embodiment of the present invention, which is described in detail below:

[0068] In step 101, the target image, the first waypoint, and the target point are acquired.

[0069] In step 102, the target image is subjected to differential processing, and the processed data is fed into an artificial neural network model for recognition to determine the positions of the first waypoint, the target point, and the guide line in the target image.

[0070] In some embodiments, step 102 includes:

[0071] The target image is desaturated to obtain a first grayscale image;

[0072] The distribution range of multiple pixel values ​​in the first grayscale image is adjusted to obtain the second grayscale image;

[0073] Perform differential and pooling operations on the second grayscale image to obtain the first data matrix;

[0074] The first data matrix is ​​fed into the artificial neural network in regions to identify the first waypoint, the target point, and the guide line;

[0075] Based on the recognition results, the positions of the first waypoint, the target point, and the guide line in the target image are extracted.

[0076] In some embodiments, adjusting the distribution range of multiple pixel values ​​in the first grayscale image to obtain the second grayscale image includes:

[0077] Multiple standard pixel value intervals are obtained, wherein each standard pixel value interval has a predetermined number of continuously distributed pixels;

[0078] Arrange multiple pixel values ​​in the first grayscale image in ascending or descending order to obtain a pixel value queue;

[0079] The pixel value queue is divided according to a predetermined number to obtain multiple first pixel value intervals;

[0080] Based on the plurality of standard pixel value ranges, the plurality of first pixel value ranges, and the first formula, the first grayscale image is adjusted to obtain a second grayscale image, wherein the first formula is:

[0081]

[0082] In the formula, For the nth pixel in the first pixel value range, It is the smallest pixel value in the nth interval of the first pixel value. SP is the largest pixel value in the nth interval of the first pixel value. nMax SP is the largest pixel value in the nth standard pixel value range. nMin It is the smallest pixel value in the nth standard pixel value range. This refers to the pixel in the adjusted nth first pixel value range.

[0083] In some implementations, performing differential and pooling operations on the second grayscale image to obtain the first data matrix includes:

[0084] Obtain the difference operator data block;

[0085] Data blocks of the same type as the difference operator data blocks are extracted multiple times from the second grayscale image according to a predetermined positional order, and used as multiple intermediate data blocks;

[0086] Calculate the dot product between the difference operator data block and the plurality of intermediate data blocks respectively to obtain multiple difference results;

[0087] Based on the position of the intermediate data block corresponding to the difference result in the second grayscale image, the multiple difference results are constructed into an intermediate data matrix;

[0088] Max pooling is performed on the intermediate data matrix to obtain the first data matrix.

[0089] In some implementations, the training process of the artificial neural network includes:

[0090] Multiple data blocks are obtained, which are obtained by performing differential and pooling operations on the grayscale image. Each data block corresponds to a waypoint label or guide line label.

[0091] The multiple data blocks are respectively input into the basic model of the artificial neural network to obtain multiple recognition outputs;

[0092] Analyze the deviation between the multiple identification outputs and the labels of the multiple data blocks;

[0093] If the deviation is greater than the recognition error threshold, then the backpropagation algorithm is used to adjust multiple parameters of the artificial neural network base model according to the deviation;

[0094] Otherwise, fix multiple parameters of the artificial neural network base model and use the artificial neural network base model as the artificial neural network.

[0095] For example, to overcome the impact of relying on guide lines and QR codes to guide the transport vehicle forward, which affects the overall transport capacity, the embodiments of the present invention use guide lines and QR codes as the main guidance methods. When obstacles obstructing passage occur, a temporary route is planned between the waypoint and the target point to avoid them. To achieve the above objective, the embodiments of the present invention acquire an image containing waypoints (locations that have already been passed), target points (the next location to be reached), and guide lines. The image is preprocessed and identified to determine the positions of the waypoints, target points, and guide lines in the image. Based on the positions in the image, the current position relative to the first waypoint and the target point is determined. Finally, the transport vehicle is guided according to the current position.

[0096] In terms of image preprocessing, the embodiments of the present invention first desaturate the image, and then adjust the pixel value range of the image. For example... Figure 2As shown in the figure, this diagram illustrates the principle of grayscale image pixel value adjustment provided by an embodiment of the present invention. Before adjustment, multiple standard pixel value intervals 201 are first obtained. In the figure, the Y-axis represents the number of pixels, and the X-axis represents the pixel value. It can be seen that the standard pixel value intervals 201 are a set of pixels distributed within a predetermined pixel value interval. Then, the grayscale image pixel values ​​are arranged, and according to the number of standard pixel value intervals 201, the arrangement is transformed into multiple first pixel value intervals 202. We can observe that each first pixel value interval 202 has a corresponding standard pixel value interval 201. The embodiment of the present invention adjusts the grayscale image using a first formula, which is:

[0097]

[0098] In the formula, For the nth pixel in the first pixel value range, It is the smallest pixel value in the nth interval of the first pixel value. SP is the largest pixel value in the nth interval of the first pixel value. nMax SP is the largest pixel value in the nth standard pixel value range. nMin It is the smallest pixel value in the nth standard pixel value range. This refers to the pixel in the adjusted nth first pixel value range.

[0099] It can be seen that the pixel value distribution range of the grayscale image after grayscale adjustment (the second grayscale image) is basically the same as that of the standard pixel value range. This results in the brightness and darkness being basically consistent with the image within the standard pixel value range. Good image preprocessing can help with image recognition.

[0100] Since the processed image is large and has a lot of data and features, the embodiment of the present invention performs differential processing on the image. Figure 3 This is a schematic diagram of the second grayscale image differential processing procedure provided by an embodiment of the present invention. First, the present invention obtains a differential operator data block. One type of differential operator data block is:

[0101]

[0102] Then, by sliding, intermediate data blocks 302 of the same type as the difference operator's data blocks are repeatedly extracted from the second grayscale image 301 and multiplied with the difference operator 303 using a dot product operation (bitwise multiplication and summation). The result is used as the difference result 304. It can be seen that each difference result 304 corresponds to an intermediate data block 302. According to the relative position of the intermediate data blocks 302 in the second grayscale image 301, the multiple difference results 304 are arranged to obtain the intermediate data matrix. The intermediate data matrix is ​​then max-pooled to obtain the first data matrix. In some application scenarios, the max-pooling method involves repeatedly sliding data blocks from the intermediate data matrix and selecting the data with the largest value to replace the extracted data block position to obtain the pooled data matrix.

[0103] The first data matrix is ​​divided into blocks and fed into an artificial neural network for recognition. The data blocks are identified based on whether they contain the first waypoint, the target point, and the guide line. Finally, the position of the data block in the first data matrix is ​​extracted as the position of the target in the target image.

[0104] Artificial neural networks are built by training on multiple data blocks. The acquisition process for these data blocks is the same as the processes and principles of image grayscale conversion, grayscale adjustment, differentiation, and pooling described above. Each of these data blocks includes a target to be identified, which may be a waypoint, a target point, or a guide line, and the included target is used as a label. These data blocks are fed into the artificial neural network, which will provide a recognition result. Before sufficient training, this recognition result has a large deviation. Based on the recognition deviation, the parameters of the artificial neural network are adjusted to gradually approach the correct recognition result. Once the parameters of the artificial neural network are fixed, it can be used as a recognition model.

[0105] In step 103, the current position is determined based on the first waypoint, the target point, and the position of the guide line in the target image.

[0106] In some embodiments, step 103 includes:

[0107] A positioning model is obtained, wherein the positioning model outputs an indication of the current position based on the first waypoint, the target point, and the position of the guide line in the target image;

[0108] The positions of the first waypoint, the target point, and the guide line in the target image are input into the positioning model to obtain the current position;

[0109] The positioning model is determined based on multiple positioning datasets. Each positioning dataset includes current location sample data, first waypoint sample data, target point sample data, and position data of guide line sample data in the sample image.

[0110] The multiple location datasets are input into the location model respectively to obtain multiple location outputs;

[0111] Based on the multiple positioning outputs and the current location sample data of the multiple positioning datasets, the positioning deviation of the positioning model is determined;

[0112] If the positioning deviation is greater than the positioning deviation threshold, then multiple parameters of the positioning model are adjusted according to the positioning deviation.

[0113] In some implementations, the positioning model is:

[0114]

[0115] In the formula, P(x) is the x-axis coordinate of the current position, mmax is the total exponent, and w xm c is the weight of the m-th x-axis. x Let P(y) be the x-axis offset constant, and P(y) be the y-axis coordinate indicating the current position. ym c is the weight of the m-th y-axis. y Let w be the y-axis bias constant, X be the joint variable, and w be the y-axis bias constant. b Let x be the weight of the b-th sample data. b This represents the position of the b-th sample image.

[0116] For example, after obtaining the positions of the first waypoint, the target point, and the guide line in the target image, the present invention can send the above data into the positioning model to determine the current position.

[0117] The localization model is as follows:

[0118]

[0119] In the formula, P(x) is the x-axis coordinate of the current position, mmax is the total exponent, and w xm c is the weight of the m-th x-axis. x Let P(y) be the x-axis offset constant, and P(y) be the y-axis coordinate indicating the current position. ym c is the weight of the m-th y-axis. y Let w be the y-axis bias constant, X be the joint variable, and w be the y-axis bias constant. b Let x be the weight of the b-th sample data. b This represents the position of the b-th sample image.

[0120] We can see that the model has multiple parameters, such as weight parameters, bias constants, etc. These parameters are determined through localization datasets, each of which includes the position data of the current position sample data, the first waypoint sample, the target point sample, and the guide line sample in the sample image.

[0121] When adjusting the parameters, firstly, the total number of parameters in the localization model is reduced or decreased to be less than and close to the number of parameters in the localization dataset. Then, the position data of the first waypoint sample, the target point sample, and the guide line sample in the sample image are fed into the localization model. If the indication given by the localization model deviates significantly from the current position sample data in the dataset, the parameters of the localization model are adjusted according to the deviation. After adjustment, the dataset is input again, and this process is repeated until the deviation is less than a threshold. At this point, the parameters of the localization model can be fixed and used as the final localization model.

[0122] As we can see, the embodiments of the present invention do not establish a model of the target's position and current position in the image through physical relationships, but instead model it through samples. The model is easy to build and the recognition effect is controllable.

[0123] In step 104, the direction of travel is adjusted according to the current position.

[0124] This invention discloses a visual guidance method for a transport vehicle. First, it acquires a target image, a first waypoint, and a target travel point. Then, it performs differential processing on the target image and feeds the processed data into an artificial neural network model for identification, determining the positions of the first waypoint, the target travel point, and the guide line within the target image. Next, based on the positions of these elements, the current position is determined. Finally, the travel direction is adjusted according to the current position. This invention, while primarily relying on guide lines for guidance, determines the current position through image recognition. Therefore, it allows the transport vehicle to deviate from the guide lines to a certain extent, thus avoiding oncoming vehicles without requiring some vehicles to wait for avoidance, improving overall transport efficiency and reducing the number of guide lines.

[0125] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0126] The following are embodiments of the apparatus of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0127] Figure 4 This is a functional block diagram of the visual guidance device for the guided transport vehicle provided in an embodiment of the present invention, with reference to... Figure 4 The guided transport vehicle visual guidance device includes: a data acquisition module 401, an image recognition module 402, a positioning module 403, and a navigation module 404, wherein:

[0128] Data acquisition module 401 is used to acquire the target image, the first waypoint, and the target point.

[0129] Image recognition module 402 is used to perform differential processing on the target image and send the processed data into an artificial neural network model for recognition to determine the positions of the first waypoint, the target point, and the guide line in the target image.

[0130] The positioning module 403 is used to determine the current position based on the position of the first waypoint, the target point, and the guide line in the target image;

[0131] The navigation module 404 is used to adjust the direction of travel according to the current position.

[0132] Figure 5 This is a functional block diagram of the electronic device provided in an embodiment of the present invention. For example... Figure 5 As shown, the electronic device 5 of this embodiment includes a processor 500 and a memory 501, wherein the memory 501 stores a computer program 502 that can run on the processor 500. When the processor 500 executes the computer program 502, it implements the steps of the various guided transport vehicle visual guidance methods and embodiments described above, for example... Figure 1 Steps 101 to 104 are shown.

[0133] For example, the computer program 502 may be divided into one or more modules / units, which are stored in the memory 501 and executed by the processor 500 to complete the present invention.

[0134] The electronic device 5 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. The electronic device 5 may include, but is not limited to, a processor 500 and a memory 501. Those skilled in the art will understand that... Figure 5 This is merely an example of electronic device 5 and does not constitute a limitation on electronic device 5. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 5 may also include input / output devices, network access devices, buses, etc.

[0135] The processor 500 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0136] The memory 501 can be an internal storage unit of the electronic device 5, such as a hard disk or memory. The memory 501 can also be an external storage device of the electronic device 5, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 501 can include both internal and external storage units of the electronic device 5. The memory 501 is used to store the computer program 502 and other programs and data required by the electronic device 5. The memory 501 can also be used to temporarily store data that has been output or will be output.

[0137] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the aforementioned method embodiments, and will not be repeated here.

[0138] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0139] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0140] In the embodiments provided by this invention, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0141] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0142] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0143] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above-described embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various methods and apparatus embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0144] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions 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 invention, and should all be included within the protection scope of the present invention.

Claims

1. A visual guidance method for a guided transport vehicle, characterized in that, include: Acquire the target image, the first waypoint, and the target point; The target image is subjected to differential processing, and the processed data is fed into an artificial neural network model for recognition to determine the positions of the first waypoint, the target point, and the guide line in the target image, including: The target image is desaturated to obtain a first grayscale image; The distribution range of multiple pixel values ​​in the first grayscale image is adjusted to obtain the second grayscale image; Perform differential and pooling operations on the second grayscale image to obtain the first data matrix; The first data matrix is ​​fed into the artificial neural network in regions to identify the first waypoint, the target point, and the guide line; Based on the recognition results, the positions of the first waypoint, the target point, and the guide line in the target image are extracted. The current position is determined based on the position of the first waypoint, the target point, and the guide line in the target image; Adjust the direction of travel based on the current position.

2. The visual guidance method for a guided transport vehicle according to claim 1, characterized in that, The step of adjusting the distribution range of multiple pixel values ​​in the first grayscale image to obtain the second grayscale image includes: Multiple standard pixel value intervals are obtained, wherein each standard pixel value interval has a predetermined number of continuously distributed pixels; Arrange multiple pixel values ​​in the first grayscale image in ascending or descending order to obtain a pixel value queue; The pixel value queue is divided according to a predetermined number to obtain multiple first pixel value intervals; Based on the plurality of standard pixel value ranges, the plurality of first pixel value ranges, and the first formula, the first grayscale image is adjusted to obtain a second grayscale image, wherein the first formula is: In the formula, For the first The pixels in the first pixel value range For the first The smallest pixel value in the first pixel value range. For the first The largest pixel value in the first pixel value range For the first The largest pixel value in a standard pixel value range For the first The smallest pixel value in a standard pixel value range. For the adjusted number The pixels in the first pixel value range.

3. The visual guidance method for a guided transport vehicle according to claim 1, characterized in that, The step of performing differential and pooling operations on the second grayscale image to obtain the first data matrix includes: Obtain the difference operator data block; Data blocks of the same type as the difference operator data blocks are extracted multiple times from the second grayscale image according to a predetermined positional order, and used as multiple intermediate data blocks; Calculate the dot product between the difference operator data block and the plurality of intermediate data blocks respectively to obtain multiple difference results; Based on the position of the intermediate data block corresponding to the difference result in the second grayscale image, the multiple difference results are constructed into an intermediate data matrix; Max pooling is performed on the intermediate data matrix to obtain the first data matrix.

4. The visual guidance method for a guided transport vehicle according to claim 1, characterized in that, The training process of the artificial neural network includes: Multiple data blocks are obtained, which are obtained by performing differential and pooling operations on the grayscale image. Each data block corresponds to a waypoint label or guide line label. The multiple data blocks are respectively input into the basic model of the artificial neural network to obtain multiple recognition outputs; Analyze the deviation between the multiple identification outputs and the labels of the multiple data blocks; If the deviation is greater than the recognition error threshold, then the backpropagation algorithm is used to adjust multiple parameters of the artificial neural network base model according to the deviation; Otherwise, fix multiple parameters of the artificial neural network base model and use the artificial neural network base model as the artificial neural network.

5. The visual guidance method for a guided transport vehicle according to any one of claims 1-4, characterized in that, Determining the current position based on the first waypoint, the target point, and the position of the guide line in the target image includes: A positioning model is obtained, wherein the positioning model outputs an indication of the current position based on the first waypoint, the target point, and the position of the guide line in the target image; The positions of the first waypoint, the target point, and the guide line in the target image are input into the positioning model to obtain the current position; The positioning model is determined based on multiple positioning datasets. Each positioning dataset includes current location sample data, first waypoint sample data, target point sample data, and position data of guide line sample data in the sample image. The multiple location datasets are input into the location model respectively to obtain multiple location outputs; Based on the multiple positioning outputs and the current location sample data of the multiple positioning datasets, the positioning deviation of the positioning model is determined; If the positioning deviation is greater than the positioning deviation threshold, then multiple parameters of the positioning model are adjusted according to the positioning deviation.

6. The visual guidance method for a guided transport vehicle according to claim 5, characterized in that, The positioning model is as follows: In the formula, Indicator of current location Axis coordinates For the total number of exponents, For the first indivual Axis weight, for Axis offset constant, Indicator of current location Axis coordinates For the first indivual Axis weight, for Axis offset constant, As joint variables, For the first Weights of individual sample data For the first Each sample image location.

7. A visual guidance device for a guided transport vehicle, characterized in that, For implementing the visual guidance method for a guided transport vehicle as described in any one of claims 1-6, the visual guidance device for the guided transport vehicle comprises: The data acquisition module is used to acquire the target image, the first waypoint, and the target point. The image recognition module is used to perform differential processing on the target image and input the processed data into an artificial neural network model for recognition, determining the positions of the first waypoint, the target point, and the guide line in the target image, including: The target image is desaturated to obtain a first grayscale image; the distribution range of multiple pixel values ​​in the first grayscale image is adjusted to obtain a second grayscale image; the second grayscale image is subjected to difference and pooling operations to obtain a first data matrix; the first data matrix is ​​fed into an artificial neural network in regions to identify the first waypoint, the target point, and the guide line; the positions of the first waypoint, the target point, and the guide line in the target image are extracted based on the identification results. The positioning module is used to determine the current position based on the position of the first waypoint, the target point, and the guide line in the target image; as well as, The navigation module is used to adjust the direction of travel based on the current location.

8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 6 above.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 6 above.

Images