An AGV self-adaptive navigation method and system

Abstract

The present application relates to the technical field of AGV car navigation, and relates to an AGV car adaptive navigation method and system, wherein a sensor array installed on the AGV car is used to monitor the environment around the AGV car in real time, sensor data is analyzed by the control system of the AGV car, the actual environmental position of the pedestrian is identified, and the actual distance between the AGV car and the pedestrian at each moving time point is determined; if the actual distance between the AGV car and the pedestrian at a certain moving time point is less than the set minimum comfortable social distance, the moving time point is recorded as a reference time point, and the obstacle avoidance path of the AGV car is automatically planned; finally, the AGV car is navigated and controlled by the control system of the AGV car, and the appropriate social distance can make people feel more comfortable when sharing space with robots, reduce people's nervousness and insecurity, and the AGV car can avoid pedestrians and reduce the risk of collision accidents when sharing the working environment with people, protect the safety of pedestrians, and improve the intelligent level of the AGV car.

Description

Technical Field

[0001] The present invention belongs to the technical field of AGV navigation and relates to an AGV self-adaptive navigation method and system. Background Art

[0002] Automated guided vehicles (AGVs) play a vital role in modern logistics, manufacturing, and many automated environments. Therefore, precise navigation control of these vehicles is not only important but also necessary. This is because the AGV's navigation system ensures that they can move safely and efficiently in complex working environments, whether in narrow warehouse corridors, busy production lines, or in spaces shared with human workers. Precise navigation control not only improves operational efficiency and reduces the time cost of material handling, but also significantly reduces the accidents and losses that may occur due to navigation errors, thereby ensuring production safety and equipment integrity.

[0003] AGVs still have some flaws and shortcomings in achieving pedestrian avoidance and maintaining a certain social distance:

[0004] Humans follow certain social rules when avoiding others, such as maintaining appropriate personal space and taking precedence. AGV systems may not fully consider these social rules when designing, causing their behavior to be perceived as unnatural or impolite by humans, leading to discomfort or confusion.

[0005] AGVs may not be able to adapt to environmental changes in a timely manner in a changing environment. This lack of adaptability may cause delays in identifying new obstacle avoidance paths, increasing the risk of collision. They may currently be inefficient in processing multiple decisions, making it difficult to maximize efficiency while ensuring safety. Summary of the Invention

[0006] In view of the above problems in the prior art, the present invention provides an AGV adaptive navigation method and system to solve the above technical problems.

[0007] In order to achieve the above-mentioned and other purposes, the technical solutions adopted by the present invention are as follows:

[0008] A first aspect of the present invention provides an AGV adaptive navigation method, the method comprising the following steps:

[0009] S1, monitor the surrounding environment of the AGV in real time through a sensor array installed on the AGV, wherein the sensor array includes a distance sensor and an image acquisition device;

[0010] S2, the AGV control system analyzes sensor data to identify the actual environmental location of the pedestrian, and then determines the actual distance between the AGV and the pedestrian at each moving time point;

[0011] S3, if it is identified that the actual distance between the AGV and the pedestrian at a certain moving time point is less than the set minimum comfortable social distance, the moving time point is recorded as a reference time point, and the obstacle avoidance path of the AGV is automatically planned;

[0012] S4, according to the result of the planned obstacle avoidance path of the AGV, the AGV is navigated and controlled by the control system of the AGV.

[0013] For example, the actual environmental position of the pedestrian is identified, and the specific identification process is as follows:

[0014] S2-1, the image acquisition device on the AGV captures the working environment image and transmits it to the control system, and the control system pre-processes the working environment image to obtain the pixel value of each coordinate point corresponding to the working environment image , i represents the horizontal coordinate of each coordinate point, j represents the vertical coordinate of each coordinate point, and the horizontal Sobel operator and the vertical Sobel operator are extracted respectively;

[0015] S2-2, the convolution calculation of each coordinate point corresponding to the working environment image and the horizontal Sobel operator is performed to obtain the horizontal gradient value of each coordinate point corresponding to the working environment image , u and v represent the index of the filter, and the convolution calculation of the working environment image and the vertical Sobel operator is performed synchronously to obtain the vertical gradient value of each coordinate point corresponding to the working environment image ;

[0016] S2-3, the gradient amplitude value of each coordinate point corresponding to the working environment image is further calculated , the set low threshold and high threshold are obtained, and the gradient amplitude value of each coordinate point corresponding to the working environment image is compared with them. The coordinate point with a gradient amplitude value higher than the high threshold is considered as an edge coordinate point, and the coordinate point with a gradient amplitude value lower than the low threshold is excluded.

[0017] S2-4, if there is a coordinate point with a gradient amplitude value between the low threshold and the high threshold, it is screened out, and the final comparison is performed. When the comparison step in the S2-3 step is determined to be finished, the coordinate point with a gradient amplitude value between the low threshold and the high threshold is continuously compared. If the gradient amplitude value of a certain coordinate point is between the low threshold and the high threshold and has a connection relationship with the determined edge coordinate point, it indicates that the coordinate point is also an edge coordinate point, otherwise the coordinate point is excluded.

[0018] S2-5, according to the judgment steps of S2-3 and S2-4, each environmental object in the working environment image is obtained.

[0019] The actual environment position of the pedestrian is identified, and the specific identification process further includes the following steps:

[0020] Each environment object in the working environment image is removed from the working environment image to obtain each environment object image in the working environment image, thereby obtaining the RGB value of each pixel point in each environment object image, and converting the RGB color space to the YCbCr color space. The conversion can be performed by the following formula:

[0021] Y = 0.299R + 0.587G + 0.114B;

[0022] Cb = -0.1687R - 0.3313G + 0.5B + 128;

[0023] Cr = 0.5R - 0.4187G - 0.0813B + 128;

[0024] Further, the YCbCr color space value of each pixel point in each environment object image is obtained, the YCbCr color space value of each pixel point in each environment object image is filled into each environment object contour, a processed image of each environment object is further obtained, and the processed image is compared with an image set corresponding to the object name stored in the database. Thus, the object name corresponding to each environment object is obtained, the object name is determined as the environmental object of the pedestrian, and the position of the pedestrian in the working environment image is further determined.

[0025] The position of the pedestrian in the working environment image is matched with the environment map, and the actual environment position of the pedestrian is finally obtained.

[0026] The actual distance between the AGV car and the pedestrian at each moving time point is determined, and the specific determination process is as follows:

[0027] The image acquisition device on the AGV car is calibrated using a checkerboard calibration method to obtain the focal length F of the image acquisition device on the AGV car.

[0028] The working environment images at each moving time point are continuously captured by the image acquisition device on the AGV car, thereby determining the actual environment position of the pedestrian at each moving time point;

[0029] The pixel distance between the top of the pedestrian's head and the bottom of the pedestrian's feet is obtained from the working environment images at each moving time point u is the number of each moving time point, u = 1, 2,..., p;

[0030] And from the working environment images at each moving time point, an environment object with a known actual height is selected as a reference object, and the pixel height of the reference object in the working environment images at each moving time point is simultaneously obtained ;

[0031] Finally, the actual distance between the AGV and the pedestrian at each moving time point is determined by using a calculation formula , wherein represents the actual height of the reference object in the working environment image at the u-th moving time point, represents the actual distance from the image acquisition device to the reference object in the working environment image at the u-th moving time point.

[0032] The specific calculation formula of the actual distance from the image acquisition device to the reference object in the working environment image at each moving time point is .

[0033] Exemplarily, the automatic planning of the obstacle avoidance path of the AGV specifically represents the automatic planning of the obstacle avoidance travel speed and the obstacle avoidance travel direction of the AGV.

[0034] Exemplarily, the automatic planning of the obstacle avoidance travel speed and the obstacle avoidance travel direction of the AGV, the specific adjustment process is as follows:

[0035] obtain the travel speed of the AGV at the current moving time point and the steering speed ;

[0036] and obtain the maximum acceleration and the maximum steering acceleration of the AGV corresponding to the set;

[0037] The moving speed set of the AGV in the straight line direction corresponding to the previous time point at the reference time point is represented as , and the steering angle set of the AGV corresponding to the previous time point at the reference time point is represented as , is the time interval difference of the reference time point corresponding to the previous time point;

[0038] The moving speed set and the steering angle set of the AGV in the straight line direction corresponding to the previous time point at the reference time point are paired with each other to obtain a speed pair set of the AGV corresponding to the previous time point at the reference time point, and the objective function of the AGV corresponding to the previous time point at the reference time point is specifically represented as follows:

[0039] ;

[0040] wherein respectively represent the set weight coefficients, and , is the Euclidean distance difference between the calculated expected position of the AGV using the speed pair set and the target transport position corresponding to the previous time point at the reference time point, a minimum distance between the calculated expected position of the AGV corresponding to the speed group pair set at the reference time point and the actual environment position of the pedestrian corresponding to the last time point, a speed optimization value of the AGV corresponding to the last time point at the reference time point, a path smoothness of the AGV corresponding to the last time point at the reference time point;

[0041] traversing the speed group pair set of the AGV corresponding to the last time point at the reference time point, calculating the target function value corresponding to each speed group pair in the speed group pair set of the AGV corresponding to the last time point at the reference time point, and then selecting the moving speed in the straight line direction in the speed group pair with the maximum target function value as the obstacle avoidance running speed of the AGV corresponding to the last time point at the reference time point;

[0042] selecting the deviation angle corresponding to the steering angle in the speed group pair with the maximum target function value as the obstacle avoidance running direction of the AGV corresponding to the last time point at the reference time point.

[0043] For example, the coordinate position (x, y) of the AGV at the reference time point in the environment map is obtained, and the coordinate position (x1, y1) of the target transportation position in the environment map is obtained. The Euclidean distance difference between the calculated expected position of the AGV corresponding to the speed group pair set at the reference time point and the target transportation position is calculated (x', y') represents the calculated expected position of the AGV corresponding to the speed group pair set at the reference time point;

[0044] The coordinate position (x2, y2) of the actual environment position of the pedestrian at the reference time point in the environment map is obtained, and a minimum distance between the calculated expected position of the AGV corresponding to the speed group pair set at the reference time point and the actual environment position of the pedestrian is calculated .

[0045] For example, the linear speed of the AGV at the reference time point is obtained , and a speed optimization value of the AGV corresponding to the last time point at the reference time point is calculated , , which represents the maximum linear speed of the AGV corresponding to the last time point at the reference time point;

[0046] The running speed and the steering speed of the AGV corresponding to the last time point at the reference time point are obtained, and a path smoothness of the AGV corresponding to the last time point at the reference time point is calculated .​

[0047] The second aspect of the present application provides an AGV self-adaptive navigation system, comprising:

[0048] A vehicle data acquisition module: a sensor array installed on the AGV real-time monitors the environment around the AGV, the sensor array comprising a distance sensor and an image acquisition device;

[0049] An actual distance calculation module: the control system of the AGV identifies the actual environmental position of the pedestrian by analyzing the sensor data, and then determines the actual distance between the AGV and the pedestrian at each moving time point;

[0050] A social path adjustment module: if the actual distance between the AGV and the pedestrian at a certain moving time point is less than the set minimum comfortable social distance, the moving time point is recorded as a reference time point, and then the obstacle avoidance path of the AGV is automatically planned;

[0051] An AGV control module: according to the planned obstacle avoidance path of the AGV, the AGV is navigated and controlled by the control system of the AGV.

[0052] Another aspect of the present application provides an AGV self-adaptive navigation device, comprising an electronic device and a computer readable storage medium;

[0053] The electronic device comprises a processor, a memory and a communication bus, and the memory stores a computer readable program executable by the processor;

[0054] The communication bus realizes the connection and communication between the processor and the memory;

[0055] The processor realizes the steps of the AGV self-adaptive navigation method as described above when executing the computer readable program;

[0056] The computer readable storage medium stores one or more programs, which can be executed by one or more processors to realize the steps of the AGV self-adaptive navigation method as described above.

[0057] As described above, the present application provides an AGV self-adaptive navigation method and system, which has at least the following beneficial effects:

[0058] The application provides an AGV self-adaptive navigation method and system, which realizes real-time monitoring of the surrounding environment of the AGV through a sensor array installed on the AGV, analyzes the sensor data by the control system of the AGV, identifies the actual environmental position of the pedestrian, and further determines the actual distance between the AGV and the pedestrian at each moving time point, and if the actual distance between the AGV and the pedestrian at a certain moving time point is less than the set minimum comfortable social distance, the moving time point is recorded as a reference time point, and the obstacle avoidance path of the AGV is automatically planned, and finally the AGV is navigated and controlled by the control system of the AGV. On the one hand, when the AGV shares the working environment with people, it can avoid pedestrians and reduce the risk of collision accidents, and protect the safety of pedestrians. Through effective avoidance strategies, the AGV can avoid pedestrians without interrupting its tasks, which can improve work efficiency and reduce unnecessary pauses or detours caused by avoiding pedestrians. On the other hand, maintaining an appropriate social distance can make people feel more comfortable when sharing space with robots, reducing people's sense of tension and insecurity, not only improving the intelligent level of the AGV, but also being a key factor in promoting the wide application of artificial intelligence and robot technology in practical applications. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0060] Figure 1 The schematic diagram of the method steps of the application is shown.

[0061] Figure 2 The schematic diagram of the system modules of the application is shown.

[0062] Figure 3 The schematic diagram of the structure of the device shown in the application is shown. DETAILED DESCRIPTION

[0063] The following will combine the above content of the application, which is only an example and description of the concept of the application. Those skilled in the art can make various modifications or supplements to the described specific embodiments or use similar ways to replace, as long as they do not deviate from the concept of the application or exceed the scope defined by the present claims, which should belong to the protection scope of the application.

[0064] Embodiment 1

[0065] Please refer to Figure 1 The AGV self-adaptive navigation method shown in the application comprises the following steps:

[0066] S1, monitor the surrounding environment of the AGV in real time through a sensor array installed on the AGV, wherein the sensor array includes a distance sensor and an image acquisition device;

[0067] S2, the AGV control system analyzes sensor data to identify the actual environmental location of the pedestrian, and then determines the actual distance between the AGV and the pedestrian at each moving time point;

[0068] Based on the above embodiment, the actual environmental location of the pedestrian is identified. The specific identification process is as follows:

[0069] S2-1. Capture the working environment image through the image acquisition device on the AGV and transmit it to the control system. The control system pre-processes the working environment image to obtain the pixel value of each coordinate point in the working environment image. , i represents the horizontal coordinate of each coordinate point, j represents the vertical coordinate of each coordinate point, and the horizontal Sobel operator is extracted respectively. and vertical Sobel operator ;

[0070] S2-2, convolve the work environment image corresponding to each coordinate point with the horizontal Sobel operator to obtain the horizontal gradient value of each coordinate point in the work environment image , u and v represent the index of the traversal filter respectively, and the working environment image is convolved with the vertical Sobel operator to obtain the vertical gradient value of each coordinate point of the working environment image ;

[0071] u and v are offsets relative to the currently processed pixel (i, j). For the 3x3 Sobel operator, u and v take values ​​in the set {-1, 0, 1}. When calculating the gradient at position (i, j), u and v will traverse the neighbors around the pixel, including itself: for example, when u=-1 and v=-1, it is processing the neighbor of the upper left corner of the pixel (i, j). and They represent the value of the Sobel operator at the position (u, v). Since u and υ are (-1, 0, 1}, 1 needs to be added when actually indexing the Sobel operator in order to correctly access the elements of the 3x3 matrix in programming languages ​​that start with 0 as the index.

[0072] S2-3. Further calculate the gradient amplitude of each coordinate point of the working environment image. , the gradient amplitude of each coordinate point in the working environment image is compared with the set low threshold and high threshold, the coordinate point with the gradient amplitude higher than the high threshold is regarded as an edge coordinate point, and the coordinate point with the gradient amplitude lower than the low threshold is excluded;

[0073] S2-4, if the gradient amplitude of a certain coordinate point is between the low threshold and the high threshold, the coordinate point is screened out, and the comparison step is continued when the comparison step in the determination end S2-3 step is determined, if the gradient amplitude of a certain coordinate point is between the low threshold and the high threshold and there is a connection relationship with the determined edge coordinate point, it is indicated that the coordinate point is also an edge coordinate point, otherwise the coordinate point is excluded;

[0074] S2-5, according to the judgment steps of S2-3 and S2-4, each environment object in the working environment image is obtained.

[0075] On the basis of the above embodiment, the actual environmental position of the pedestrian is identified, and the specific identification process further includes the following steps:

[0076] Each environment object in the working environment image is excluded from the working environment image to obtain each environment object image in the working environment image, thereby obtaining the RGB value of each pixel point in each environment object image, and converting the RGB color space to YCbCr color space, which can be converted by the following formula:

[0077] Y=0.299R+0.587G+0.114B;

[0078] Cb=-0.1687R-0.3313G+0.5B+128;

[0079] Cr=0.5R-0.4187G-0.0813B+128;

[0080] Further, the YCbCr color space value of each pixel point in each environment object image is obtained, the YCbCr color space value of each pixel point in each environment object image is filled into each environment object contour, a processing image of each environment object is further obtained, and the processing image is compared with the image set corresponding to the object name stored in the database, thereby comparing the object name corresponding to each environment object, screening out the object name determined as the environmental object of the pedestrian, and further determining the position of the pedestrian in the working environment image;

[0081] The position of the pedestrian in the working environment image is matched with the environmental map, and the actual environmental position of the pedestrian is finally obtained.

[0082] On the basis of the above embodiment, the actual distance between the AGV car and the pedestrian at each moving time point is determined, and the specific determination process is:

[0083] Use the checkerboard calibration method to calibrate the image acquisition device on the AGV to obtain the focal length F of the image acquisition device on the AGV;

[0084] The image acquisition device on the AGV continuously captures images of the working environment at each moving time point, thereby determining the actual environmental position of the pedestrian at each moving time point;

[0085] Obtain the pixel distance between the top of the pedestrian's head and the bottom of his feet from the work environment image at each moving time point , u is the number of each moving time point, u=1,2,...p;

[0086] And select an environmental object with a known actual height from the working environment image at each moving time point as a reference object, and synchronously obtain the pixel height of the reference object in the working environment image at each moving time point ;

[0087] It should be added that the specific process of obtaining environmental objects with known actual heights is as follows:

[0088] Obtain the object name corresponding to each environmental object in the working environment image at each moving time point, and compare it with the actual height corresponding to each object name stored in the database. The actual height of each environmental object in the working environment image at each moving time point is obtained by comparison, and any environmental object is arbitrarily selected as the environmental object with the actual height.

[0089] Finally, the calculation formula is used to determine the actual distance between the AGV and the pedestrian at each moving time point. ,in represents the actual height of the reference object in the working environment image at the u-th moving time point, represents the actual distance from the image acquisition device to the reference object in the working environment image at the u-th moving time point;

[0090] The specific calculation formula for the actual distance from the image acquisition device to the reference object in the working environment image at each moving time point is: .

[0091] S3. If the actual distance between the AGV and the pedestrian at a certain moving time point is identified to be less than the set minimum comfortable social distance, the moving time point is recorded as a reference time point, and the obstacle avoidance path of the AGV is automatically planned;

[0092] On the basis of the above embodiment, the automatic planning of the obstacle avoidance path of the AGV is specifically manifested in automatically planning the obstacle avoidance driving speed and obstacle avoidance driving direction of the AGV.

[0093] Based on the above embodiment, the obstacle avoidance speed and direction of the AGV are automatically planned. The specific adjustment process is as follows:

[0094] Get the driving speed of the AGV at the current moving time and steering speed ;

[0095] And obtain the maximum acceleration corresponding to the AGV car and maximum steering acceleration ;

[0096] Then the set of moving speeds of the AGV car in the straight line direction corresponding to the previous time point at the reference time point is expressed as , the steering angle set of the AGV car at the reference time point corresponding to the previous time point is expressed as , The time interval difference between the reference time point and the previous time point;

[0097] The AGV car's linear moving speed set and steering angle set at the reference time point corresponding to the previous time point are paired in pairs to obtain the AGV car's speed pair set at the reference time point corresponding to the previous time point. The objective function of the AGV car at the reference time point corresponding to the previous time point is specifically expressed as follows:

[0098] ;

[0099] in Represent the set weight coefficients, and , The Euclidean distance difference between the expected position calculated by the speed pair set and the target transport position of the AGV at the reference time point corresponding to the previous time point, The minimum distance between the expected position calculated by the AGV using the speed pair set at the reference time point corresponding to the previous time point and the actual position of the pedestrian in the environment, is the optimized speed value of the AGV at the reference time point corresponding to the previous time point, is the path smoothness of the AGV at the reference time point corresponding to the previous time point;

[0100] Traverse the set of speed pairs of the AGV at the reference time point corresponding to the previous time point, calculate the corresponding objective function value of each speed pair in the speed pair set of the AGV at the reference time point corresponding to the previous time point, and then select the linear moving speed in the speed pair with the largest objective function value as the obstacle avoidance driving speed of the AGV at the reference time point corresponding to the previous time point;

[0101] The deflection angle corresponding to the steering angle in the speed group with the largest objective function value is selected as the obstacle avoidance driving direction of the AGV at the reference time point corresponding to the previous time point.

[0102] What needs to be explained is that by traversing different and The AGV can evaluate its performance at different speeds and steering angles at the next time point. This evaluation is based on a predefined objective function that takes into account factors such as the efficiency of the AGV moving to the target, the safe distance from pedestrians, and the smoothness of the path. By selecting the optimal objective function, the AGV can evaluate its performance at different speeds and steering angles at the next time point. and , the AGV is able to adjust its travel direction and speed in real time to avoid hitting pedestrians and navigate smoothly to the target location.

[0103] Based on the above embodiment, the coordinate position (x, y) of the AGV position at the reference time point in the environment map is obtained, and the coordinate position (x1, y1) of the target transport position in the environment map is obtained. The speed pair set of the AGV car at the reference time point corresponding to the previous time point is used to calculate the Euclidean distance difference between the expected position calculated by the speed pair set at the reference time point corresponding to the previous time point and the target transport position. , (x', y') represents the expected position of the AGV at the reference time point corresponding to the previous time point using the speed pair set;

[0104] Obtain the coordinate position (x2, y2) of the actual environmental position of the pedestrian at the reference time point in the environmental map, and calculate the minimum distance between the expected position calculated by the AGV at the reference time point corresponding to the previous time point using the speed group pair set and the actual environmental position of the pedestrian .

[0105] Based on the above embodiment, the linear speed of the AGV at the reference time point is obtained. , calculate the optimal speed value of the AGV car at the reference time point corresponding to the previous time point , Indicates the maximum linear speed of the AGV car;

[0106] Get the driving speed of the AGV at the reference time point corresponding to the previous time point and steering speed , calculate the path smoothness of the AGV car at the reference time point corresponding to the previous time point .

[0107] S4. Based on the obstacle avoidance path result of the planned AGV, the AGV is navigated and controlled by the AGV control system.

[0108] Embodiment 2

[0109] Please refer to Figure 2 As shown in the figure, an AGV self-adaptive navigation system, the system comprises:

[0110] Car data acquisition module: through the sensor array installed on the AGV car real-time monitoring AGV car around the environment, the sensor array includes distance sensor and image acquisition device;

[0111] Actual distance calculation module: AGV car control system by analyzing sensor data, identify the actual environment of the pedestrian position, and then determine the actual distance between AGV car and the pedestrian at each moving time point;

[0112] Social path adjustment module: if the actual distance between AGV car and the pedestrian at a certain moving time point is less than the set minimum comfortable social distance, the moving time point is recorded as the reference time point, and the AGV car obstacle avoidance path is automatically planned;

[0113] AGV car control module: according to the AGV car obstacle avoidance path result obtained by planning, the AGV car is navigated and controlled by the AGV car control system.

[0114] Embodiment 3

[0115] According to the AGV self-adaptive navigation device shown in an example embodiment, comprising: electronic equipment and computer readable storage medium;

[0116] Figure 3 It is a structure schematic diagram of an AGV self-adaptive navigation device provided by the embodiment of the application, which can have great difference due to different configuration or performance, and can include one or more processors (Central Processing Units, CPU) and one or more memories, wherein the memory stores at least one computer program, the at least one computer program is loaded and executed by the processor to realize the above-mentioned AGV self-adaptive navigation method.

[0117] The electronic equipment includes a processor, a memory and a communication bus, the memory stores a computer readable program which can be executed by the processor;

[0118] The communication bus realizes the connection communication between the processor and the memory;

[0119] The processor realizes the above-mentioned AGV self-adaptive navigation method by calling the computer program stored in the memory and executing the computer readable program;

[0120] The computer readable storage medium stores one or more programs for execution by one or more processors of the electronic device, and the one or more programs, when executed, provide a user input interface to implement the AGV adaptive navigation method described above.

[0121] It should be understood that the size of the sequence number of the above processes does not mean the order of execution in various embodiments of the present application. The execution order of the processes should be determined according to their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0122] It should be understood that determining B according to A does not mean that B is determined only according to A, but B can also be determined according to A and / or other information.

[0123] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0124] Finally, the above is only a preferred embodiment of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. An AGV adaptive navigation method, characterized in that: The steps include: S1, monitor the surrounding environment of the AGV in real time through a sensor array installed on the AGV, wherein the sensor array includes a distance sensor and an image acquisition device; S2, the AGV control system analyzes sensor data to identify the actual environmental location of the pedestrian, and then determines the actual distance between the AGV and the pedestrian at each moving time point; S3. If the actual distance between the AGV and the pedestrian at a certain moving time point is identified to be less than the set minimum comfortable social distance, the moving time point is recorded as a reference time point, and the obstacle avoidance path of the AGV is automatically planned; S4. Based on the obstacle avoidance path result of the planned AGV, the AGV is navigated and controlled by the AGV control system; Automatically planning the obstacle avoidance path of the AGV is specifically manifested in automatically planning the obstacle avoidance driving speed and direction of the AGV. The specific adjustment process is as follows: Get the driving speed of the AGV at the current moving time and steering speed ; And obtain the maximum driving acceleration corresponding to the AGV car and maximum turning speed ; Then the set of AGV vehicle's speed in the straight line direction corresponding to the previous time point at the reference time point is expressed as , the turning speed set of the AGV car at the reference time point corresponding to the previous time point is expressed as , The time interval difference between the reference time point and the previous time point; The AGV car's linear driving speed set and turning speed set at the reference time point corresponding to the previous time point are paired in pairs to obtain the speed pair set of the AGV car at the reference time point corresponding to the previous time point. The objective function of the AGV car at the reference time point corresponding to the previous time point is specifically expressed as follows: ； in Represent the set weight coefficients, and , The Euclidean distance difference between the expected position calculated by the speed pair set and the target transport position of the AGV at the reference time point corresponding to the previous time point, The minimum distance between the expected position calculated by the AGV using the speed pair set at the reference time point corresponding to the previous time point and the actual position of the pedestrian in the environment, is the optimized speed value of the AGV at the reference time point corresponding to the previous time point, is the path smoothness of the AGV at the reference time point corresponding to the previous time point; Traverse the set of speed pairs of the AGV at the reference time point corresponding to the previous time point, calculate the corresponding objective function value of each speed pair in the speed pair set of the AGV at the reference time point corresponding to the previous time point, and then select the straight-line driving speed in the speed pair with the largest objective function value as the obstacle avoidance driving speed of the AGV at the reference time point corresponding to the previous time point; The deflection angle corresponding to the steering speed in the speed group with the largest objective function value is selected as the obstacle avoidance driving direction of the AGV at the reference time point corresponding to the previous time point.

2. The AGV adaptive navigation method according to claim 1, characterized in that: Determine the actual distance between the AGV and the pedestrian at each moving time point. The specific determination process is as follows: Use the checkerboard calibration method to calibrate the image acquisition device on the AGV to obtain the focal length F of the image acquisition device on the AGV; The image acquisition device on the AGV continuously captures images of the working environment at each moving time point, thereby determining the actual environmental position of the pedestrian at each moving time point; Obtain the pixel distance between the top of the pedestrian's head and the bottom of his feet from the work environment image at each moving time point , u is the number of each moving time point, u=1,2,...p; And select an environmental object with a known actual height from the working environment image at each moving time point as a reference object, and synchronously obtain the pixel height of the reference object in the working environment image at each moving time point ; Finally, the calculation formula is used to determine the actual distance between the AGV and the pedestrian at each moving time point. ,in represents the actual height of the reference object in the working environment image at the u-th moving time point, represents the actual distance from the image acquisition device to the reference object in the working environment image at the u-th moving time point; The specific calculation formula for the actual distance from the image acquisition device to the reference object in the working environment image at each moving time point is: .

3. The AGV adaptive navigation method according to claim 1, characterized in that: Get the coordinate position (x, y) of the AGV car at the reference time point in the environment map, and get the coordinate position (x1, y1) of the target transportation position in the environment map. Use the speed pair set of the AGV car at the reference time point corresponding to the previous time point to calculate the Euclidean distance difference between the expected position calculated by the speed pair set at the reference time point corresponding to the previous time point and the target transportation position. , (x', y') represents the expected position of the AGV at the reference time point corresponding to the previous time point using the speed pair set; Obtain the coordinate position (x2, y2) of the actual environmental position of the pedestrian at the reference time point in the environmental map, and calculate the minimum distance between the expected position calculated by the AGV at the reference time point corresponding to the previous time point using the speed group pair set and the actual environmental position of the pedestrian .

4. The AGV adaptive navigation method according to claim 3, characterized in that: Get the linear speed of the AGV at the reference time point , calculate the optimal speed value of the AGV car at the reference time point corresponding to the previous time point , Indicates the maximum linear speed of the AGV car; Get the driving speed of the AGV at the reference time point corresponding to the previous time point and steering speed , calculate the path smoothness of the AGV car at the reference time point corresponding to the previous time point .

5. An AGV adaptive navigation system, characterized in that: The method is based on the AGV adaptive navigation method described in any one of claims 1 to 4 and includes the following modules: Car data acquisition module: monitors the surrounding environment of the AGV in real time through the sensor array installed on the AGV, which includes a distance sensor and an image acquisition device; Actual distance calculation module: The AGV control system analyzes sensor data to identify the actual environmental location of pedestrians, and then determines the actual distance between the AGV and the pedestrian at each moving time point; Social path adjustment module: If the actual distance between the AGV and the pedestrian at a certain moving time point is identified to be less than the set minimum comfortable social distance, the moving time point is recorded as the reference time point, and the obstacle avoidance path of the AGV is automatically planned; AGV control module: Based on the planned obstacle avoidance path results of the AGV, the AGV is navigated and controlled through the AGV control system.

6. An AGV adaptive navigation device, characterized by: This includes electronic equipment; The electronic device includes a processor, a memory and a communication bus, wherein the memory stores a computer-readable program that can be executed by the processor; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, the steps of the AGV adaptive navigation method as described in any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the AGV adaptive navigation method as described in any one of claims 1 to 4.

Images