AGV robot operation control method and system based on machine vision

Abstract

The invention discloses an AGV robot operation control method and system based on machine vision, and relates to the technical field of AGV robots. Comprising the steps of obtaining working road information of an AGV robot working environment, and dividing the working road information into a plurality of area parts; determining the positions of all AGV robots; determining an initial congestion index of each area of the operation road; identifying road obstacles in the working road; determining an operation walking priority index of each area in the operation road; determining at least one walking route; calculating a priority value of each walking route; and screening out the walking route with the maximum priority value as the optimal walking route. The AGV robot movement control method has the advantages that when AGV robot movement control is carried out, overall analysis of the overall working environment is achieved, the walking fluency of the AGV robot is guaranteed, and the working efficiency of the AGV robot is maximized.

Description

Technical Field

[0001] The present invention relates to the technical field of AGV robots, and in particular to an AGV robot operation control method and system based on machine vision. Background Art

[0002] The main function of AGV robots is to carry out automated mobile work. AGV robots automatically move to the target location through special landmark navigation to carry out their work. In existing industrial automation designs, AGV robots are an important means of transporting between different production processes. In existing industrial designs, landmark navigation roads for AGV robots to walk on are usually marked in advance in the workshop. AGV robots carry out material flow by walking on the navigation roads.

[0003] However, in the prior art, although AGV robots generally have 3D visual image acquisition modules, the 3D visual image acquisition of existing AGV robots is only used for path obstacle avoidance, and it is difficult to achieve an overall analysis of the AGV robot's working environment on the road. When performing mobile control of the AGV robot, there is a lack of overall analysis, which affects the walking smoothness of the AGV robot and makes it difficult to maximize the work efficiency of the AGV robot. Summary of the invention

[0004] In order to solve the above technical problems, a machine vision-based AGV robot operation control method and system are provided. This technical solution solves the problem in the above-mentioned prior art that, although AGV robots generally have a 3D visual image acquisition module, the 3D visual image acquisition of existing AGV robots is only used for path obstacle avoidance, and it is difficult to achieve an overall analysis of the AGV robot's working environment on the road. When performing the mobile control of the AGV robot, there is a lack of overall analysis, which affects the walking smoothness of the AGV robot and makes it difficult to maximize the work efficiency of the AGV robot.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A machine vision-based AGV robot operation control method, comprising:

[0007] Acquire the working road information of the working environment of the AGV robot, and divide the working road of the working environment of the AGV robot into several area parts;

[0008] Based on the UWB positioning system, determine the location of all AGV robots in the working area;

[0009] Based on the positions of all AGV robots in the working area, determine the initial congestion index of each area of ​​the working road;

[0010] Based on the working road images collected by the visual devices of all AGV robots, road obstacles in the working road are identified;

[0011] Determine the operation travel priority index of each area on the operation road based on the road obstacles on the operation road and the initial congestion index of each area on the operation road;

[0012] Determine an initial position of the AGV robot and a target position of the AGV robot, and determine at least one walking route based on the working road information of the working environment of the AGV robot;

[0013] Based on the operation walking priority index of each area in the operation road, the priority value of each walking route is calculated;

[0014] The walking route with the largest priority value is selected as the optimal walking route, and a control instruction is issued to control the AGV robot to walk along the optimal walking route.

[0015] Preferably, the UWB positioning system is used to determine the positions of all AGV robots in the working area, including:

[0016] The AGV robot receives the UWB signal sent by the UWB positioning base station through the built-in UWB communication module, and solves the signal data contained in the UWB signal. At the same time, based on the solved signal data, it returns the position information data to the UWB positioning base station through the UWB signal;

[0017] Based on the time difference between the sending time of the UWB signal and the receiving time of the UWB signal returned by the AGV robot, combined with the propagation speed of the UWB signal, the distance between the AGV robot and the UWB positioning base station is determined;

[0018] Determine the distance between the AGV robot and at least three UWB positioning base stations;

[0019] Draw a circle with the position of the UWB positioning base station as the center and the distance between the AGV robot and the UWB positioning base station as the radius, which is recorded as the positioning circle;

[0020] Determine the intersection of the positioning circles made by at least three UWB positioning base stations, and the intersection is the position of the AGV robot.

[0021] Preferably, the determining of the initial congestion index of each area of ​​the working road based on the positions of all AGV robots in the working area specifically includes:

[0022] Determine the road area of ​​each area of ​​the working road;

[0023] Obtain information about all AGV robots in each area, and determine the total area occupied by all AGV robots in each area based on the information about all AGV robots in each area;

[0024] Calculate the ratio of the total area occupied by all AGV robots in the area to the road area of ​​the area as the benchmark congestion index of the area;

[0025] Calculate the initial congestion index of each area of ​​the working road through the congestion calculation formula;

[0026] The congestion calculation formula is specifically:

[0027]

[0028] Where D j is the initial congestion index of the jth area in the working road, N j is the benchmark congestion index in the jth area of ​​the working road, N max is the maximum value of the baseline congestion index of all areas, N min is the minimum value of the baseline congestion index of all regions, and m is the total number of regions.

[0029] Preferably, the identifying of road obstacles in the working road based on the working road images collected by the visual devices of all AGV robots specifically includes:

[0030] Acquire the collected working road image through visible light vision, and extract the red, green and blue color components of the pixel points in the working road image;

[0031] Obtain the weighted coefficients of the human eye sensitivity of red, green and blue, substitute the red, green and blue components of each pixel of the road image into the weighted average formula, and obtain the grayscale of each pixel of the working road image;

[0032] Determine the standard grayscale interval of the working road;

[0033] The working road image area outside the standard grayscale range of the working road image pixel points is recorded as an obstacle area;

[0034] By using infrared vision, it is determined whether there is a three-dimensional structure in the obstacle area. If so, it is determined that there is an obstacle in the obstacle area. If not, it is determined that there is no obstacle in the obstacle area.

[0035] Summarize the locations of obstacles in the process of all AGV robots, and determine whether at least two AGV robots recognize obstacles at the same location. If so, determine that there is a fixed obstacle at the location; if not, determine that there is no fixed obstacle at the location;

[0036] The weighted average formula is specifically:

[0037] Gray(a,b)=α1R(a,b)+α2G(a,b)+α3B(a,b),

[0038] Where Gray(a,b) is the grayscale of the pixel in the ath row and bth column in the working road image, α1, α2, and α3 are the weighted coefficients of the human eye sensitivity to red, green, and blue, respectively, and R(a,b), G(a,b), and B(a,b) are the red, green, and blue components of the pixel in the ath row and bth column, respectively.

[0039] Preferably, the determining of the working travel priority index of each area in the working road based on the road obstacles in the working road and the initial congestion index of each area in the working road specifically includes:

[0040] Based on the road obstacles in the working road, determine the number of AGV robots that can be accommodated in parallel on each road section in the area;

[0041] Based on the maximum number of AGV robots that can be accommodated in parallel on each road section in the area and the total length of the working roads in the area, the road parallel fitting index of the area is calculated by the road fitting formula;

[0042] Based on the regional road parallel fitting index and the initial congestion index of the region, the operation walking priority index of each region is calculated through a comprehensive evaluation formula;

[0043] The road fitting formula is specifically:

[0044]

[0045] In the road fitting formula, R is the road parallel fitting index of the region, i is the total length of the road section in the region that can accommodate a maximum of i AGV robots in parallel, and s i is the maximum value of the maximum number of AGV robots that can be accommodated in parallel on each road section in the area, and S is the total length of the working road in the area;

[0046] The comprehensive evaluation formula is specifically:

[0047]

[0048] Where Y j is the priority index of the operation walking in the jth area, R j is the road parallel fitting index of the jth region, β1 is the weight emphasizing the road attribute, and β2 is the weight emphasizing the congestion attribute.

[0049] Preferably, the method of determining the number of AGV robots that can be accommodated in parallel in each road section in the area based on the road obstacles in the working road specifically includes:

[0050] The difference between the walking road width at the location of the fixed obstacle in the calculation area and the road width occupied by the fixed obstacle is recorded as the actual walking road width at the location;

[0051] Get the actual walking road width at each road section in the area;

[0052] Divide the actual walking road width by the walking width of the AGV robot and round down to get the number of AGV robots that can be accommodated in parallel on the road section in the area.

[0053] Preferably, the step of calculating the priority value of each walking route based on the work walking priority index of each area in the work road specifically includes:

[0054] Determine the length of each area of ​​the walking route;

[0055] The walking index of the walking route in the area is obtained by multiplying the length of the area through which the walking route passes by the operation walking priority index of the area;

[0056] The walking indices of all areas of the walking route are accumulated and summed to obtain the priority value of the walking route.

[0057] Furthermore, the present invention proposes an AGV robot operation control system based on machine vision, which is used to implement the above-mentioned AGV robot operation control method based on machine vision, including:

[0058] A region division module, wherein the region division module is used to obtain the working road information of the working environment of the AGV robot and divide the working road of the working environment of the AGV robot into several region parts;

[0059] A positioning module, the positioning module is electrically connected to the area division module, and the positioning module is used to determine the positions of all AGV robots in the working area based on the UWB positioning system, and determine the initial congestion index of each area of ​​the working road based on the positions of all AGV robots in the working area;

[0060] An obstacle recognition module, which is used to identify road obstacles in the working road based on the working road images collected by the visual devices of all AGV robots;

[0061] A road state analysis module, the road state analysis module is electrically connected to the area division module, the positioning module and the obstacle identification module, and the road state analysis module is used to determine the operation travel priority index of each area in the operation road based on the road obstacles in the operation road and the initial congestion index of each area in the operation road;

[0062] A route planning module, the route planning module is electrically connected to the road state analysis module and the area division module, and is used to determine the initial position of the AGV robot and the target position of the AGV robot, determine at least one walking route based on the working road information of the working environment of the AGV robot, and calculate the priority value of each walking route based on the working walking priority index of each area in the working road, screen out the walking route with the largest priority value as the optimal walking route, and issue a control instruction to control the AGV robot to walk along the optimal walking route.

[0063] Optionally, the positioning module includes:

[0064] UEB system, the UEB system includes at least three UWB positioning base stations and a UWB communication module built into the AGV robot. The AGV robot receives the UWB signal sent by the UWB positioning base station through the built-in UWB communication module, and solves the signal data contained in the UWB signal. At the same time, based on the solved signal data, the position information data is returned to the UWB positioning base station through the UWB signal;

[0065] A signal analysis unit, the signal analysis unit is used to determine the distance between the AGV robot and the UWB positioning base station based on the time difference between the transmission time of the UWB signal and the reception time of the UWB signal returned by the AGV robot, combined with the propagation speed of the UWB signal;

[0066] The position determination unit is used to draw a circle with the position of the UWB positioning base station as the center and the distance between the AGV robot and the UWB positioning base station as the radius, which is recorded as the positioning circle, and determine the intersection of the positioning circles made by at least three UWB positioning base stations. The intersection is the position of the AGV robot.

[0067] Optionally, the route planning module includes:

[0068] A start-end planning unit, which is used to determine the initial position of the AGV robot and the target position of the AGV robot, and determine at least one walking route based on the working road information of the working environment of the AGV robot;

[0069] A route analysis and calculation unit, the route analysis and calculation unit is used to calculate the priority value of each walking route based on the work walking priority index of each area in the work road;

[0070] The route determination unit is used to select a walking route with the largest priority value as the optimal walking route, issue a control instruction, and control the AGV robot to walk along the optimal walking route.

[0071] Compared with the prior art, the present invention has the following beneficial effects:

[0072] The present invention proposes an AGV robot operation control scheme based on machine vision. The vision module is used to analyze the collected data of the specific road surface information of the navigation road when the robot is walking. The UEB positioning system is used to analyze the aggregation state of all AGV robots in the working environment and the movement properties of the AGV robots in the working environment. Then, the movement of the AGV robot is comprehensively controlled in combination with the movement road data in the working environment. This method can effectively improve the adaptability of the AGV robot when moving in the working environment, thereby effectively improving the efficiency of the AGV robot when working. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 This is a flow chart of the AGV robot operation control method based on machine vision proposed by the present invention;

[0074] Figure 2 A flow chart of a method for determining the positions of all AGV robots in a working area in the present invention;

[0075] Figure 3 A flow chart of a method for determining an initial congestion index of each area of ​​a working road in the present invention;

[0076] Figure 4 A flow chart of a method for identifying road obstacles on a working road in the present invention;

[0077] Figure 5 A flow chart of a method for determining the operation travel priority index of each area in the operation road in the present invention;

[0078] Figure 6 A flow chart of a method for determining the number of AGV robots that can be accommodated in parallel in each road section in an area in the present invention;

[0079] Figure 7 The figure is a flow chart of the method for calculating the priority value of each walking route in the present invention. DETAILED DESCRIPTION

[0080] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0081] Reference Figure 1 As shown, a machine vision-based AGV robot operation control method includes:

[0082] Acquire the working road information of the working environment of the AGV robot, and divide the working road of the working environment of the AGV robot into several area parts;

[0083] Based on the UWB positioning system, determine the location of all AGV robots in the working area;

[0084] Based on the positions of all AGV robots in the working area, determine the initial congestion index of each area of ​​the working road;

[0085] Based on the working road images collected by the visual devices of all AGV robots, road obstacles in the working road are identified;

[0086] Determine the operation travel priority index of each area on the operation road based on the road obstacles on the operation road and the initial congestion index of each area on the operation road;

[0087] Determine an initial position of the AGV robot and a target position of the AGV robot, and determine at least one walking route based on the working road information of the working environment of the AGV robot;

[0088] Based on the operation walking priority index of each area in the operation road, the priority value of each walking route is calculated;

[0089] The walking route with the largest priority value is selected as the optimal walking route, and a control instruction is issued to control the AGV robot to walk along the optimal walking route.

[0090] This solution uses the visual module to analyze the collected data of the specific road surface information of the navigation road when the robot is walking, and combines the UEB positioning system to analyze the aggregation status of all AGV robots in the working environment and the mobile properties of the AGV robots in the working environment, and then combines the mobile road data in the working environment to comprehensively control the movement of the AGV robot.

[0091] Reference Figure 2 As shown, the UWB positioning system is used to determine the positions of all AGV robots in the working area, including:

[0092] The AGV robot receives the UWB signal sent by the UWB positioning base station through the built-in UWB communication module, and solves the signal data contained in the UWB signal. At the same time, based on the solved signal data, it returns the position information data to the UWB positioning base station through the UWB signal;

[0093] Based on the time difference between the sending time of the UWB signal and the receiving time of the UWB signal returned by the AGV robot, combined with the propagation speed of the UWB signal, the distance between the AGV robot and the UWB positioning base station is determined;

[0094] Determine the distance between the AGV robot and at least three UWB positioning base stations;

[0095] Draw a circle with the position of the UWB positioning base station as the center and the distance between the AGV robot and the UWB positioning base station as the radius, which is recorded as the positioning circle;

[0096] Determine the intersection of the positioning circles made by at least three UWB positioning base stations, and the intersection is the position of the AGV robot.

[0097] This solution uses UWB communication positioning technology to perform high-precision positioning of all AGV robots in the working environment and can achieve high-precision positioning and distance measurement, thereby achieving high-precision positioning of AGV robots.

[0098] Reference Figure 3 As shown, the determination of the initial congestion index of each area of ​​the working road based on the positions of all AGV robots in the working area specifically includes:

[0099] Determine the road area of ​​each area of ​​the working road;

[0100] Obtain information about all AGV robots in each area, and determine the total area occupied by all AGV robots in each area based on the information about all AGV robots in each area;

[0101] Calculate the ratio of the total area occupied by all AGV robots in the area to the road area of ​​the area as the benchmark congestion index of the area;

[0102] Calculate the initial congestion index of each area of ​​the working road through the congestion calculation formula;

[0103] The congestion calculation formula is specifically:

[0104]

[0105] Where D j is the initial congestion index of the jth area in the working road, N j is the benchmark congestion index in the jth area of ​​the working road, N max is the maximum value of the baseline congestion index of all areas, N min is the minimum value of the baseline congestion index of all regions, and m is the total number of regions.

[0106] Since there are multiple AGV robots working and moving simultaneously in the workshop, this solution uses UWB technology to locate the position of each AGV robot in real time, and measures the road congestion status by the number of AGV robots in each area of ​​the working environment.

[0107] Reference Figure 4 As shown, the identification of road obstacles in the working road based on the working road images collected by the visual devices of all AGV robots specifically includes:

[0108] Acquire the collected working road image through visible light vision, and extract the red, green and blue color components of the pixel points in the working road image;

[0109] Obtain the weighted coefficients of the human eye sensitivity of red, green and blue, substitute the red, green and blue components of each pixel of the road image into the weighted average formula, and obtain the grayscale of each pixel of the working road image;

[0110] Determine the standard grayscale interval of the working road;

[0111] The working road image area outside the standard grayscale range of the working road image pixel points is recorded as an obstacle area;

[0112] By using infrared vision, it is determined whether there is a three-dimensional structure in the obstacle area. If so, it is determined that there is an obstacle in the obstacle area. If not, it is determined that there is no obstacle in the obstacle area.

[0113] Summarize the locations of obstacles in the process of all AGV robots, and determine whether at least two AGV robots recognize obstacles at the same location. If so, determine that there is a fixed obstacle at the location; if not, determine that there is no fixed obstacle at the location;

[0114] The weighted average formula is specifically:

[0115] Gray(a,b)=α1R(a,b)+α2G(a,b)+α3B(a,b),

[0116] Where Gray(a,b) is the grayscale of the pixel in the ath row and bth column in the working road image, α1, α2, and α3 are the weighted coefficients of the human eye sensitivity to red, green, and blue, respectively, and R(a,b), G(a,b), and B(a,b) are the red, green, and blue components of the pixel in the ath row and bth column, respectively.

[0117] It is understandable that in the working environment of the AGV robot, the moving path of the AGV robot is usually marked with special landmark navigation signs, that is, the working road of the AGV robot. In this solution, the areas in the working road of the AGV robot that do not meet the special landmark navigation signs are identified through visible light recognition. At the same time, infrared vision is used to determine whether there are three-dimensional obstacles in the area that hinder the movement of the AGV robot, and the road status is evaluated in combination with the visual recognition results of multiple AGV robots.

[0118] Reference Figure 5 As shown, the determination of the working travel priority index of each area in the working road based on the road obstacles in the working road and the initial congestion index of each area in the working road specifically includes:

[0119] Based on the road obstacles in the working road, determine the number of AGV robots that can be accommodated in parallel on each road section in the area;

[0120] Based on the maximum number of AGV robots that can be accommodated in parallel on each road section in the area and the total length of the working roads in the area, the road parallel fitting index of the area is calculated by the road fitting formula;

[0121] Based on the regional road parallel fitting index and the initial congestion index of the region, the operation walking priority index of each region is calculated through a comprehensive evaluation formula;

[0122] The road fitting formula is specifically:

[0123]

[0124] In the road fitting formula, R is the road parallel fitting index of the region, i is the total length of the road section in the region that can accommodate a maximum of i AGV robots in parallel, and s i is the maximum value of the maximum number of AGV robots that can be accommodated in parallel on each road section in the area, and S is the total length of the working road in the area;

[0125] The comprehensive evaluation formula is specifically:

[0126]

[0127] Where Y j is the priority index of the operation walking in the jth area, R j is the road parallel fitting index of the jth region, β1 is the weight emphasizing the road attribute, and β2 is the weight emphasizing the congestion attribute.

[0128] Reference Figure 6 As shown, the method of determining the number of AGV robots that can be accommodated in parallel in each road section in the area based on the road obstacles in the working road specifically includes:

[0129] The difference between the walking road width at the location of the fixed obstacle in the calculation area and the road width occupied by the fixed obstacle is recorded as the actual walking road width at the location;

[0130] Get the actual walking road width at each road section in the area;

[0131] Divide the actual walking road width by the walking width of the AGV robot and round down to get the number of AGV robots that can be accommodated in parallel on the road section in the area.

[0132] Reference Figure 7As shown, the calculation of the priority value of each walking route based on the work walking priority index of each area in the work road specifically includes:

[0133] Determine the length of each area of ​​the walking route;

[0134] The walking index of the walking route in the area is obtained by multiplying the length of the area through which the walking route passes by the operation walking priority index of the area;

[0135] The walking indices of all areas of the walking route are accumulated and summed to obtain the priority value of the walking route.

[0136] It is understandable that when planning roads, it is necessary to avoid areas with obstacles and large numbers of AGV robots gathered on the roads as much as possible. Based on this, this solution will subtract the length of the path through each area and the operation walking priority index of each area. The operation walking priority index represents the obstacles and the gathering of AGV robots in each area, which determines the walking speed of the AGV robot. The length of the path represents the length of the walking path, which determines the walking distance of the AGV robot. By comprehensively evaluating the priority value of each path, the optimal walking control of the AGV robot can be achieved.

[0137] Further, based on the same inventive concept as the above-mentioned AGV robot operation control method based on machine vision, this solution proposes an AGV robot operation control system based on machine vision, including:

[0138] A region division module, wherein the region division module is used to obtain the working road information of the working environment of the AGV robot and divide the working road of the working environment of the AGV robot into several region parts;

[0139] A positioning module, the positioning module is electrically connected to the area division module, and the positioning module is used to determine the positions of all AGV robots in the working area based on the UWB positioning system, and determine the initial congestion index of each area of ​​the working road based on the positions of all AGV robots in the working area;

[0140] An obstacle recognition module, which is used to identify road obstacles in the working road based on the working road images collected by the visual devices of all AGV robots;

[0141] A road state analysis module, the road state analysis module is electrically connected to the area division module, the positioning module and the obstacle identification module, and the road state analysis module is used to determine the operation travel priority index of each area in the operation road based on the road obstacles in the operation road and the initial congestion index of each area in the operation road;

[0142] A route planning module, the route planning module is electrically connected to the road state analysis module and the area division module, and is used to determine the initial position of the AGV robot and the target position of the AGV robot, determine at least one walking route based on the working road information of the working environment of the AGV robot, and calculate the priority value of each walking route based on the working walking priority index of each area in the working road, screen out the walking route with the largest priority value as the optimal walking route, and issue a control instruction to control the AGV robot to walk along the optimal walking route.

[0143] The positioning module comprises:

[0144] UEB system, the UEB system includes at least three UWB positioning base stations and a UWB communication module built into the AGV robot. The AGV robot receives the UWB signal sent by the UWB positioning base station through the built-in UWB communication module, and solves the signal data contained in the UWB signal. At the same time, based on the solved signal data, the position information data is returned to the UWB positioning base station through the UWB signal;

[0145] A signal analysis unit, the signal analysis unit is used to determine the distance between the AGV robot and the UWB positioning base station based on the time difference between the transmission time of the UWB signal and the reception time of the UWB signal returned by the AGV robot, combined with the propagation speed of the UWB signal;

[0146] The position determination unit is used to draw a circle with the position of the UWB positioning base station as the center and the distance between the AGV robot and the UWB positioning base station as the radius, which is recorded as the positioning circle, and determine the intersection of the positioning circles made by at least three UWB positioning base stations. The intersection is the position of the AGV robot.

[0147] The route planning module includes:

[0148] A start-end planning unit, which is used to determine the initial position of the AGV robot and the target position of the AGV robot, and determine at least one walking route based on the working road information of the working environment of the AGV robot;

[0149] A route analysis and calculation unit, the route analysis and calculation unit is used to calculate the priority value of each walking route based on the work walking priority index of each area in the work road;

[0150] The route determination unit is used to select a walking route with the largest priority value as the optimal walking route, issue a control instruction, and control the AGV robot to walk along the optimal walking route.

[0151] The process of using the above system is:

[0152] Step 1: The area division module obtains the working road information of the AGV robot working environment and divides the working road of the AGV robot working environment into several area parts;

[0153] Step 2: The positioning module calls the UEB system. The AGV robot receives the UWB signal sent by the UWB positioning base station through the built-in UWB communication module, and solves the signal data contained in the UWB signal. At the same time, based on the solved signal data, the position information data is returned to the UWB positioning base station through the UWB signal. Then, the signal analysis unit determines the distance between the AGV robot and the UWB positioning base station based on the time difference between the sending time of the UWB signal and the receiving time of the UWB signal returned by the AGV robot, combined with the propagation speed of the UWB signal. Then, the position determination unit is used to draw a circle with the position of the UWB positioning base station as the center and the distance between the AGV robot and the UWB positioning base station as the radius, recorded as the positioning circle, and determine the intersection of the positioning circles made by at least three UWB positioning base stations. The intersection is the position of the AGV robot. The positioning module determines the initial congestion index of each area of ​​the working road based on the positions of all AGV robots in the working area;

[0154] Step 3: The obstacle recognition module identifies road obstacles on the working road based on the working road images collected by the visual devices of all AGV robots;

[0155] Step 4: The road state analysis module determines the operation travel priority index of each area in the operation road based on the road obstacles in the operation road and the initial congestion index of each area in the operation road;

[0156] Step 5: The route planning module determines the initial position and target position of the AGV robot, determines at least one walking route based on the working road information of the AGV robot's working environment, and calculates the priority value of each walking route based on the working walking priority index of each area in the working road. The walking route with the largest priority value is selected as the optimal walking route, and a control instruction is issued to control the AGV robot to walk along the optimal walking route.

[0157] In summary, the advantages of the present invention are: when performing the movement control of the AGV robot, a comprehensive analysis of the overall working environment is achieved, the walking smoothness of the AGV robot is ensured, and the working efficiency of the AGV robot is maximized.

[0158] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A machine vision-based AGV robot operation control method, characterized in that: include: Acquire the working road information of the working environment of the AGV robot, and divide the working road of the working environment of the AGV robot into several area parts; Based on the UWB positioning system, determine the location of all AGV robots in the working area; Based on the positions of all AGV robots in the working area, determine the initial congestion index of each area of ​​the working road; Based on the working road images collected by the visual devices of all AGV robots, road obstacles in the working road are identified; Determine the operation travel priority index of each area on the operation road based on the road obstacles on the operation road and the initial congestion index of each area on the operation road; Determine an initial position of the AGV robot and a target position of the AGV robot, and determine at least one walking route based on the working road information of the working environment of the AGV robot; Based on the operation walking priority index of each area in the operation road, the priority value of each walking route is calculated; The walking route with the largest priority value is selected as the optimal walking route, and a control instruction is issued to control the AGV robot to walk along the optimal walking route.

2. The AGV robot operation control method based on machine vision according to claim 1 is characterized in that: The UWB positioning system is used to determine the positions of all AGV robots in the working area, including: The AGV robot receives the UWB signal sent by the UWB positioning base station through the built-in UWB communication module, and solves the signal data contained in the UWB signal. At the same time, based on the solved signal data, it returns the position information data to the UWB positioning base station through the UWB signal; Based on the time difference between the sending time of the UWB signal and the receiving time of the UWB signal returned by the AGV robot, combined with the propagation speed of the UWB signal, the distance between the AGV robot and the UWB positioning base station is determined; Determine the distance between the AGV robot and at least three UWB positioning base stations; Draw a circle with the position of the UWB positioning base station as the center and the distance between the AGV robot and the UWB positioning base station as the radius, which is recorded as the positioning circle; Determine the intersection of the positioning circles made by at least three UWB positioning base stations, and the intersection is the position of the AGV robot.

3. The AGV robot operation control method based on machine vision according to claim 2 is characterized in that: The determination of the initial congestion index of each area of ​​the working road based on the positions of all AGV robots in the working area specifically includes: Determine the road area of ​​each area of ​​the working road; Obtain information about all AGV robots in each area, and determine the total area occupied by all AGV robots in each area based on the information about all AGV robots in each area; Calculate the ratio of the total area occupied by all AGV robots in the area to the road area of ​​the area as the benchmark congestion index of the area; Calculate the initial congestion index of each area of ​​the working road through the congestion calculation formula; The congestion calculation formula is specifically: Where D j is the initial congestion index of the jth area in the working road, N j is the benchmark congestion index in the jth area of ​​the working road, N max is the maximum value of the baseline congestion index of all areas, N min is the minimum value of the baseline congestion index of all regions, and m is the total number of regions.

4. The AGV robot operation control method based on machine vision according to claim 3 is characterized in that: The identification of road obstacles in the working road based on the working road images collected by the visual devices of all AGV robots specifically includes: Acquire the collected working road image through visible light vision, and extract the red, green and blue color components of the pixel points in the working road image; Obtain the weighted coefficients of the human eye sensitivity of red, green and blue, substitute the red, green and blue components of each pixel of the road image into the weighted average formula, and obtain the grayscale of each pixel of the working road image; Determine the standard grayscale interval of the working road; The working road image area outside the standard grayscale range of the working road image pixel points is recorded as an obstacle area; By using infrared vision, it is determined whether there is a three-dimensional structure in the obstacle area. If so, it is determined that there is an obstacle in the obstacle area. If not, it is determined that there is no obstacle in the obstacle area. Summarize the locations of obstacles in the process of all AGV robots, and determine whether at least two AGV robots recognize obstacles at the same location. If so, determine that there is a fixed obstacle at the location; if not, determine that there is no fixed obstacle at the location; The weighted average formula is specifically: Gray(a,b)=α1R(a,b)+α2G(a,b)+α3B(a,b), Where Gray(a,b) is the grayscale of the pixel in the ath row and bth column in the working road image, α1, α2, and α3 are the weighted coefficients of the human eye sensitivity to red, green, and blue, respectively, and R(a,b), G(a,b), and B(a,b) are the red, green, and blue components of the pixel in the ath row and bth column, respectively.

5. The AGV robot operation control method based on machine vision according to claim 4 is characterized in that: The determining of the operation travel priority index of each area in the operation road based on the road obstacles in the operation road and the initial congestion index of each area in the operation road specifically includes: Based on the road obstacles in the working road, determine the number of AGV robots that can be accommodated in parallel on each road section in the area; Based on the maximum number of AGV robots that can be accommodated in parallel on each road section in the area and the total length of the working roads in the area, the road parallel fitting index of the area is calculated by the road fitting formula; Based on the regional road parallel fitting index and the initial congestion index of the region, the operation walking priority index of each region is calculated through a comprehensive evaluation formula; The road fitting formula is specifically: In the road fitting formula, R is the road parallel fitting index of the region, i is the total length of the road section in the region that can accommodate a maximum of i AGV robots in parallel, and s i is the maximum value of the maximum number of AGV robots that can be accommodated in parallel on each road section in the area, and S is the total length of the working road in the area; The comprehensive evaluation formula is specifically: Where Y j is the priority index of the operation walking in the jth area, R j is the road parallel fitting index of the jth region, β1 is the weight emphasizing the road attribute, and β2 is the weight emphasizing the congestion attribute.

6. The AGV robot operation control method based on machine vision according to claim 5 is characterized in that: The method of determining the number of AGV robots that can be accommodated in parallel in each road section in the area based on the road obstacles in the working road specifically includes: The difference between the walking road width at the location of the fixed obstacle in the calculation area and the road width occupied by the fixed obstacle is recorded as the actual walking road width at the location; Get the actual walking road width at each road section in the area; Divide the actual walking road width by the walking width of the AGV robot and round down to get the number of AGV robots that can be accommodated in parallel on the road section in the area.

7. The AGV robot operation control method based on machine vision according to claim 6 is characterized in that: The calculation of the priority value of each walking route based on the work walking priority index of each area in the work road specifically includes: Determine the length of each area of ​​the walking route; The walking index of the walking route in the area is obtained by multiplying the length of the area through which the walking route passes by the operation walking priority index of the area; The walking indices of all areas of the walking route are accumulated and summed to obtain the priority value of the walking route.

8. An AGV robot operation control system based on machine vision, characterized in that: The method for controlling an AGV robot operation based on machine vision according to any one of claims 1 to 7 comprises: A region division module, wherein the region division module is used to obtain the working road information of the working environment of the AGV robot and divide the working road of the working environment of the AGV robot into several region parts; A positioning module, the positioning module is electrically connected to the area division module, and the positioning module is used to determine the positions of all AGV robots in the working area based on the UWB positioning system, and determine the initial congestion index of each area of ​​the working road based on the positions of all AGV robots in the working area; An obstacle recognition module, which is used to identify road obstacles in the working road based on the working road images collected by the visual devices of all AGV robots; A road state analysis module, the road state analysis module is electrically connected to the area division module, the positioning module and the obstacle identification module, and the road state analysis module is used to determine the operation travel priority index of each area in the operation road based on the road obstacles in the operation road and the initial congestion index of each area in the operation road; A route planning module, the route planning module is electrically connected to the road state analysis module and the area division module, and is used to determine the initial position of the AGV robot and the target position of the AGV robot, determine at least one walking route based on the working road information of the working environment of the AGV robot, and calculate the priority value of each walking route based on the working walking priority index of each area in the working road, screen out the walking route with the largest priority value as the optimal walking route, and issue a control instruction to control the AGV robot to walk along the optimal walking route.

9. The AGV robot operation control system based on machine vision according to claim 8, characterized in that: The positioning module comprises: UEB system, the UEB system includes at least three UWB positioning base stations and a UWB communication module built into the AGV robot. The AGV robot receives the UWB signal sent by the UWB positioning base station through the built-in UWB communication module, and solves the signal data contained in the UWB signal. At the same time, based on the solved signal data, the position information data is returned to the UWB positioning base station through the UWB signal; A signal analysis unit, the signal analysis unit is used to determine the distance between the AGV robot and the UWB positioning base station based on the time difference between the transmission time of the UWB signal and the reception time of the UWB signal returned by the AGV robot, combined with the propagation speed of the UWB signal; The position determination unit is used to draw a circle with the position of the UWB positioning base station as the center and the distance between the AGV robot and the UWB positioning base station as the radius, which is recorded as the positioning circle, and determine the intersection of the positioning circles made by at least three UWB positioning base stations. The intersection is the position of the AGV robot.

10. The AGV robot operation control system based on machine vision according to claim 9, characterized in that: The route planning module includes: A start-end planning unit, which is used to determine the initial position of the AGV robot and the target position of the AGV robot, and determine at least one walking route based on the working road information of the working environment of the AGV robot; A route analysis and calculation unit, the route analysis and calculation unit is used to calculate the priority value of each walking route based on the work walking priority index of each area in the work road; The route determination unit is used to select a walking route with the largest priority value as the optimal walking route, issue a control instruction, and control the AGV robot to walk along the optimal walking route.

Images