AGV intelligent loading of large components and precise delivery system to designated storage location

Abstract

The application discloses an AGV intelligent loading and designated storage location accurate delivery system for large parts, relates to the field of part transportation, and solves the problem of low efficiency of the existing automatic transportation method for large parts, and comprises a part loading module: loading height collection and analysis are performed on the large parts to obtain vertical loading adaptive parts, loading space coverage analysis and loading angle analysis are performed on the vertical loading adaptive parts and AGV to-be-loaded carriers, a part preset loading angle is obtained, intelligent loading is performed on target adaptive parts according to the part preset loading angle, part loading carriers are obtained, a path analysis module: the part loading carriers are sorted according to the passing path distance, and the candidate passing paths are analyzed according to the sorting order, and straight road passing property analysis and curved road passing property analysis are performed, target passing paths are obtained, and a delivery confirmation module: the part loading carriers are intelligently transported and delivery accuracy is confirmed according to the target passing paths, and the application can effectively improve the part loading efficiency and transportation safety.

Description

Technical Field

[0001] This invention belongs to the field of component transportation and relates to intelligent control technology, specifically an AGV intelligent loading and precise delivery system for large components to designated storage locations. Background Technology

[0002] Existing AGV intelligent loading and transportation technology has the following drawbacks when loading and transporting large components:

[0003] 1. In the current practical application scenarios of AGV intelligent loading and transportation technology, the technical standards for loading large components are generally relatively broad. The focus is mainly on ensuring that large components can successfully complete the loading action and meet the basic loading requirements. However, there is a lack of sufficient attention and effective means for the fine adjustment of the loading angle of large components. This situation is particularly prominent in the transportation of large components. Due to the failure to optimize the loading angle, the transportation space is not fully utilized, resulting in a waste of transportation resources.

[0004] 2. In the current practical application scenarios of AGV intelligent loading and transportation technology, its path planning method is usually relatively simple and direct. It often selects the shortest path based solely on the geometric travel distance, without fully considering the actual traffic conditions at the time of travel, such as the possibility that the road surface may be blocked by obstacles. As a result, the selected path is difficult to take into account various factors in actual travel, lacking comprehensiveness and accuracy, which may affect the transportation efficiency and safety of AGV.

[0005] To address this, we propose an AGV intelligent loading and precise delivery system for large components to designated storage locations. Summary of the Invention

[0006] In view of the shortcomings of existing technologies, the purpose of this invention is to provide an AGV intelligent loading and precise delivery system for large components to designated storage locations, aiming to improve the work efficiency of AGV intelligent loading and transportation, and effectively ensure the safety of transportation.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: an AGV intelligent loading and precise delivery system for large components to designated storage locations, the specific working process of each module is as follows:

[0008] Component loading module: The module collects and analyzes the loading height of large components to obtain the vertical loading height of the components. It then performs a height analysis between the vertical loading height of the components and the vertical loading limit of the vehicle. Based on the analysis results, it selects vertical load-adaptive components and performs a loading space coverage analysis between the vertical load-adaptive components and the AGV to be loaded. Based on the analysis results, it determines the target adaptable components and performs a loading angle analysis to obtain the preset loading angle of the components. Based on the preset loading angle of the components, it intelligently loads the target adaptable components to obtain the component loading vehicle.

[0009] Path analysis module: Sort the distance of the transport path for the component loading vehicle, and perform straight road trafficability analysis and curve trafficability analysis on the candidate transport paths in turn according to the sorting order, and obtain the target transport path based on the analysis results;

[0010] Delivery confirmation module: Intelligently transports the component loading vehicle according to the target travel path, and performs accurate delivery confirmation of the target adapter component after transportation.

[0011] Furthermore, the component loading vehicle is acquired, specifically as follows:

[0012] The AGV vehicle currently in the loading state is acquired to obtain the AGV to be loaded. The maximum loading height corresponding to the AGV to be loaded is obtained to obtain the vertical loading limit of the vehicle.

[0013] Large components that need to be transported by AGV vehicles are acquired, and the maximum ground clearance of each large component is obtained to obtain the component's vertical height. The ground clearance of the AGV's carrying device is also acquired to obtain the carrying device's ground clearance. The sum of the component's vertical height and the carrying device's ground clearance is calculated to obtain the component's vertical loading height. Large components whose vertical loading height is less than the vehicle's vertical loading limit are marked as vertical load adaptable components, and a sample vertical load component is randomly selected from the acquired vertical load adaptable components.

[0014] The loading space coverage of the sample vertical load component and the AGV to be loaded is analyzed. Based on the analysis results, the loading space coverage and the preset loading angle of the component are obtained.

[0015] Obtain the loading space coverage corresponding to each vertical load adapter component, compare the values ​​of the multiple obtained loading space coverage, and set the vertical load adapter component corresponding to the maximum loading space coverage as the target adapter component.

[0016] The preset loading angle of the component corresponding to the target adapter component is obtained. The target adapter component is loaded using an AGV vehicle. During the loading process, the reference point of the vertical component is kept coincident with the geometric center point of the transport device plane. The angle between the positive long axis of the vertical component and the positive half axis of the x-axis of the transport plane coordinate system is the preset loading angle of the component. The AGV vehicle that has completed the loading of the target adapter component is set as the component loading vehicle.

[0017] Furthermore, the coverage of the loading space and the preset loading angle of the components are obtained, as follows:

[0018] Planar radiometric images of the sample vertical load component are acquired to obtain a top view image of the component. The edge contour of the sample vertical load component is marked in the top view image to obtain the image region of the vertical load component. An outer circle is created in the image region of the vertical load component to obtain the outer circle of the vertical load component. The center of the outer circle of the vertical load component is marked as the reference point of the vertical load component.

[0019] The major axis of the sample vertically loaded component is identified, and the positive major axis of the vertical component is obtained from the identification result.

[0020] The AGV to be loaded is subjected to planar image acquisition of the transport device, and a plane rectangular coordinate system is created in the transport device plane to obtain the transport plane coordinate system;

[0021] The top view image layer of the component is translated to the plane of the transport device to keep the reference point of the vertical load component coincide with the origin of the coordinate system of the transport plane;

[0022] In the plane of the carrier device, the edge contour of the sample vertical load component is divided into several edge pixels, and the edge pixels that coincide with the x-axis are marked as the initial edge pixels;

[0023] The edge contour of the transport device in the top view image of the component is marked to obtain the marked area of ​​the transport device. The origin of the coordinate system and the initial edge pixel are obtained to obtain the initial point connection line. The positive long axis of the vertical component is aligned with the initial point connection line. If the marked area of ​​the transport device can cover the image area of ​​the vertical load component, the initial edge pixel is divided into effective loading pixel. If the marked area of ​​the transport device cannot cover the image area of ​​the vertical load component, the initial edge pixel is divided into invalid loading pixel.

[0024] Repeat the process of classifying the initial edge pixels, classify each edge pixel separately, and when the edge point of the positive long axis of the vertical component coincides with the effective loading pixel, calculate the ratio of the area value of the vertical component image region to the area value of the marked region of the transport device to obtain the loading plane space ratio corresponding to the effective loading pixel, and compare the values ​​of the multiple loading plane space ratios obtained, and set the loading plane space ratio with the smallest value as the loading space coverage corresponding to the sample vertical component;

[0025] The effective loading pixel corresponding to the minimum loading plane space ratio is set as the loading feature pixel. The loading feature pixel is connected to the origin of the coordinate system to set the loading feature mark line. The angle between the loading feature mark line and the positive half axis of the x-axis at the origin of the coordinate system is obtained to obtain the preset loading angle of the component.

[0026] Furthermore, the positive major axis of the vertical component is obtained, as follows:

[0027] A circumferential pixel window is set for the outer circle of the vertical load component, and a feature circumferential pixel point is arbitrarily selected on the outer circle of the vertical load component. When the circumferential pixel window is at the feature circumferential pixel point, a diameter of the outer circle of the vertical load component passing through the circumferential pixel window is drawn, and the two intersection points of the drawn diameter and the edge of the image area of ​​the vertical load component are obtained to obtain the first diameter edge intersection point and the second diameter edge intersection point. The distance between the first diameter edge intersection point and the second diameter edge intersection point is obtained to obtain the component diameter interception distance corresponding to the feature circumferential pixel point.

[0028] Using a circular traversal pixel window, starting from the feature circular pixel point, traverse in reverse half of the circumference of the vertical component to obtain the component diameter intercept distance corresponding to each circular pixel point. Compare the obtained component diameter intercept distances and set the component diameter intercept distance with the largest value as the major axis distance of the vertical component.

[0029] The first diameter edge intersection point and the second diameter edge intersection point corresponding to the major axis distance of the vertical load component are obtained. The distance value between the first diameter edge intersection point and the reference point of the vertical load component is obtained to obtain the first intersection point intercept distance. The distance value between the second diameter edge intersection point and the reference point of the vertical load component is obtained to obtain the second intersection point intercept distance. If the first intersection point intercept distance is greater than or equal to the second intersection point intercept distance, the line connecting the first diameter edge intersection point and the reference point of the vertical load component is set as the positive major axis of the vertical load component. If the first intersection point intercept distance is less than the second intersection point intercept distance, the line connecting the second diameter edge intersection point and the reference point of the vertical load component is set as the positive major axis of the vertical load component.

[0030] Furthermore, the target travel path is obtained, as follows:

[0031] Obtain the component loading vehicle, collect the regional map of the transportation area where the component loading vehicle is located, obtain the vehicle transportation area map, mark the current position of the component loading vehicle as the vehicle initial position on the vehicle transportation area map, mark the designated storage location of the target adapter component as the vehicle target position, obtain the passage path between the vehicle initial position and the vehicle target position, obtain multiple AVG vehicle passage paths, and mark the obtained AVG vehicle passage paths as candidate passage paths T1 to candidate passage paths Ta according to the path length from shortest to longest;

[0032] Perform passability analysis on candidate T1 routes, and identify abnormal passage areas based on the analysis results;

[0033] If there is no abnormal passage area in candidate path T1, then candidate path T1 is selected as the target passage path. If there is an abnormal passage area in candidate path T1, then the passageability analysis is performed on candidate path T2, and so on, until the target passage path is selected.

[0034] Furthermore, abnormal passage areas are identified for candidate passage paths T1, as follows:

[0035] The candidate T1 route is divided into straight sub-paths and curved sub-paths, and real-time planar image acquisition is performed on each sub-path to obtain multiple sub-path planar images.

[0036] Based on the sub-path planar image, the vehicle passability of the straight sub-path is analyzed for component loading, and the region type of the straight sub-path is divided according to the analysis results;

[0037] Based on the sub-path planar image, the component loading and vehicle passability of the curved sub-path is analyzed, and the region type of the curved sub-path is divided according to the analysis results.

[0038] Furthermore, the straight sub-paths are further divided into region types, as follows:

[0039] In the sub-path planar image, the two sides of the straight sub-path are marked respectively to obtain the side edge of the first straight path and the side edge of the second straight path.

[0040] Obstacle recognition is performed based on the sub-path planar image. If there is an obstacle in the sub-path planar image, the obstacle outline is marked on the sub-path planar image to obtain the obstacle marking area. Edge pixels are extracted from the obstacle marking area. Perpendicular lines to the side of the first straight path are drawn through the edge pixels. The intersection of the drawn perpendicular lines and the side of the first straight path is marked as the intersection of the first side perpendicular lines. A straight line perpendicular to the side of the second straight path is drawn through the intersection of the first side perpendicular lines to obtain multiple obstacle-covering perpendicular lines.

[0041] The number of intersections between the edge of the obstacle marking area and the obstacle coverage vertical line is counted to obtain the number of obstacle coverage intersections. If the number of obstacle coverage intersections is equal to 1, the coverage length of the obstacle marking area to the obstacle coverage vertical line is numerically collected, and the difference between the obstacle coverage vertical line length value and the coverage length value is calculated to obtain the remaining passage distance corresponding to the obstacle coverage vertical line.

[0042] If the number of intersection points covered by obstacles is greater than 1, the intersection points of the obstacle marking area and the obstacle covering vertical line are collected to obtain multiple obstacle feature intersection points. The obstacle covering vertical line is divided into multiple obstacle sub-vertical lines using the obstacle feature intersection points. The length values ​​of the obstacle sub-vertical lines not covered by the obstacle marking area are collected, and the minimum length value is set as the remaining passage distance corresponding to the obstacle covering vertical line.

[0043] The remaining passage distances corresponding to the vertical lines covered by different obstacles are compared numerically, and the remaining passage distance with the smallest value is marked as the path passage limit corresponding to the straight sub-path.

[0044] The required passage width of the component loading vehicle is numerically collected to obtain the required passage width of the vehicle. If the passage limit of the straight sub-path is less than or equal to the required passage width of the vehicle, the corresponding straight sub-area is marked as an abnormal passage area. If the passage limit of the straight sub-path is greater than the required passage width of the vehicle, the corresponding straight sub-area is marked as a normal passage area.

[0045] Furthermore, the curve sub-paths are divided into region types, as follows:

[0046] In the sub-path planar image, the edges of the straight and curved paths on both sides are marked respectively to obtain the side of the first curved path and the side of the second curved path.

[0047] If there are no obstacles in the curved sub-path in the sub-path planar image, then the curve radius of the curved sub-path is analyzed to obtain the curve radius of the sub-path.

[0048] If there are obstacles in the curved sub-path in the sub-path planar image, then the curve radius of the curved sub-path is analyzed to obtain the curve radius of the sub-path.

[0049] The required turning radius of the component loading vehicle is numerically collected to obtain the required turning radius of the vehicle. If the turning limit corresponding to the curve sub-path is less than or equal to the required turning radius of the vehicle, the corresponding straight sub-area is marked as an abnormal passage area. If the turning limit corresponding to the straight sub-path is greater than the required turning radius of the vehicle, the corresponding straight sub-area is marked as a normal passage area.

[0050] Furthermore, the curve radius of the sub-path is obtained, as follows:

[0051] In the sub-path planar image, the side of the first curve path is discretized into several side pixel points. Tangents are drawn to the side of the first curve path using the side pixel points as tangent points to obtain the side pixel tangents. Perpendiculars are drawn to the side pixel tangents through the side pixel points to obtain the side tangent point perpendicular lines. The intersection of the side tangent point perpendicular lines and the side of the first curve path is marked as the side perpendicular intersection pixel point. The midpoint of the line connecting each side pixel point and the corresponding side perpendicular intersection pixel point is collected to obtain multiple side connection midpoints. The side connection midpoints are connected sequentially by curves to obtain the sub-path curve feature curve.

[0052] Mark the start and end points of the curves in the sub-path curve feature curves respectively. Connect the start and end points of the curves to obtain the curve path feature line. Collect the length values ​​of the curve path feature straight lines to obtain the starting straight line distance of the curve. Obtain the midpoint of the sub-path curve feature curve to obtain the center point of the curve curve. Draw a perpendicular line from the center point of the curve curve to the curve path feature line to obtain the vertical distance of the curve high point.

[0053] The sub-path curve radius is obtained by calculating the starting straight distance of the curve and the vertical distance from the highest point of the curve.

[0054] Furthermore, the curve radius of the sub-path is obtained, as follows:

[0055] In the sub-path planar image, obstacles are marked to obtain the path obstacle region. The edge of the path obstacle region is discretized into several obstacle edge pixels. The minimum line distance between each obstacle edge pixel and the side of the curve is obtained to obtain multiple obstacle curve side distances. The obstacle edge pixel corresponding to the maximum obstacle curve side distance is marked as the feature edge pixel, and the side of the curve corresponding to the maximum obstacle curve side distance is marked as the feature curve side.

[0056] The side of the feature curve is discretized into several feature side pixels. Then, a sample side pixel is randomly selected from the obtained side pixels. The sample side pixel is used as the tangent point to draw the tangent line of the feature curve side. The perpendicular line from the sample side pixel to the obtained tangent line is drawn to obtain the sample pixel tangent line. In the sample pixel tangent line, the pixels whose distance from the sample side pixel is the maximum obstacle curve side distance are marked to obtain the obstacle offset pixel corresponding to the sample side pixel.

[0057] Obtain the obstacle offset pixel corresponding to each feature side pixel, and connect the obstacle offset pixels in adjacent states using a smooth curve to obtain the offset curve side corresponding to the feature curve side.

[0058] The curve radius of the sub-path curve is obtained by taking the curve side of the feature curve and the curve side of the offset curve.

[0059] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0060] 1. This invention collects and analyzes the loading height of large components to obtain the vertical loading height of the components. It then analyzes the vertical loading height of the components against the vertical loading limit of the carrier. Based on the analysis results, it selects vertically compatible components and performs a loading space coverage analysis between the vertically compatible components and the AGV to be loaded. Based on the analysis results, it determines the target compatible components and performs a loading angle analysis to obtain the preset loading angle of the components. Based on the preset loading angle of the components, it intelligently loads the target compatible components, which can make full use of the transportation space and further improve the transportation efficiency of AGV transportation equipment.

[0061] 2. This invention sorts the distance of the transport path for the component loading vehicle, and performs straight-line trafficability analysis and curve trafficability analysis on the candidate transport paths in turn according to the sorting order. Based on the analysis results, the target transport path is obtained, which can effectively screen out the shortest and most trafficable path, thereby effectively improving the safety of transporting large objects. Attached Figure Description

[0062] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0063] Figure 1 This is an overall system block diagram of the present invention;

[0064] Figure 2 This is a schematic diagram of the outer circle of the vertical load component in this invention;

[0065] Figure 3 This is a schematic diagram of the transport plane coordinate system in this invention. Detailed Implementation

[0066] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0067] Example 1

[0068] Please see Figure 1 This invention provides a technical solution: an AGV intelligent loading and precise delivery system for large components to designated storage locations. The specific processes of each module are as follows:

[0069] The component loading module collects and analyzes the loading height of large components to obtain the vertical loading height of the components. It then performs a height analysis between the vertical loading height of the components and the vertical loading limit of the vehicle. Based on the analysis results, it selects vertical load-adaptive components and performs a loading space coverage analysis between the vertical load-adaptive components and the AGV to be loaded. Based on the analysis results, it determines the target adaptable components and performs a loading angle analysis to obtain the preset loading angle of the components. Based on the preset loading angle of the components, it intelligently loads the target adaptable components to obtain the component loading vehicle.

[0070] Specifically as follows:

[0071] The AGV vehicle currently in the loading state is acquired to obtain the AGV to be loaded. The maximum loading height corresponding to the AGV to be loaded is obtained to obtain the vertical loading limit of the vehicle.

[0072] It should be noted here that:

[0073] In this application, the AGV loading device referred to herein is specifically an automated guided vehicle waiting to be loaded.

[0074] Large components that need to be transported by AGV vehicles are acquired, and the maximum ground clearance of each large component is obtained to obtain the component's vertical height. The ground clearance of the AGV's carrying device is also acquired to obtain the carrying device's ground clearance. The sum of the component's vertical height and the carrying device's ground clearance is calculated to obtain the component's vertical loading height. Large components whose vertical loading height is less than the vehicle's vertical loading limit are marked as vertical load adaptable components, and a sample vertical load component is randomly selected from the acquired vertical load adaptable components.

[0075] It should be noted here that:

[0076] In this application, the large components referred to herein are specifically large, irregularly shaped mechanical parts that cannot be stacked longitudinally;

[0077] In this application, the transport device referred to herein is specifically the physical device used for loading goods in the AGV loading vehicle;

[0078] The loading space coverage of the sample vertical load component and the AGV to be loaded is analyzed. Based on the analysis results, the loading space coverage and the preset loading angle of the component are obtained.

[0079] Specifically as follows:

[0080] Planar radiometric images of the sample vertical load component are acquired to obtain a top view image of the component. The edge contour of the sample vertical load component is marked in the top view image to obtain the image region of the vertical load component. An outer circle is created in the image region of the vertical load component to obtain the outer circle of the vertical load component. The center of the outer circle of the vertical load component is marked as the reference point of the vertical load component.

[0081] Please see Figure 2 A circumferential pixel window is set for the outer circle of the vertical load component, and a feature circumferential pixel point is arbitrarily selected on the outer circle of the vertical load component. When the circumferential pixel window is at the feature circumferential pixel point, a diameter of the outer circle of the vertical load component passing through the circumferential pixel window is drawn, and the two intersection points of the drawn diameter and the edge of the image area of ​​the vertical load component are obtained to obtain the first diameter edge intersection point and the second diameter edge intersection point. The distance between the first diameter edge intersection point and the second diameter edge intersection point is obtained to obtain the component diameter interception distance corresponding to the feature circumferential pixel point.

[0082] Using a circular traversal pixel window, starting from the feature circular pixel point, traverse in reverse half of the circumference of the vertical component to obtain the component diameter intercept distance corresponding to each circular pixel point. Compare the obtained component diameter intercept distances and set the component diameter intercept distance with the largest value as the major axis distance of the vertical component.

[0083] It should be noted here that:

[0084] In this application, since the circumcircle of the vertical load component is symmetrical, it is only necessary to traverse half of the circumcircle of the vertical load component in reverse.

[0085] The first diameter edge intersection point and the second diameter edge intersection point corresponding to the major axis distance of the vertical load component are obtained. The distance value between the first diameter edge intersection point and the reference point of the vertical load component is obtained to obtain the first intersection point intercept distance. The distance value between the second diameter edge intersection point and the reference point of the vertical load component is obtained to obtain the second intersection point intercept distance. If the first intersection point intercept distance is greater than or equal to the second intersection point intercept distance, the line connecting the first diameter edge intersection point and the reference point of the vertical load component is set as the positive major axis of the component. If the first intersection point intercept distance is less than the second intersection point intercept distance, the line connecting the second diameter edge intersection point and the reference point of the vertical load component is set as the positive major axis of the component.

[0086] The AGV to be loaded is subjected to planar image acquisition of the transport device, and a plane rectangular coordinate system is created in the transport device plane to obtain the transport plane coordinate system;

[0087] It should be noted here that:

[0088] In this application, the origin of the carrier plane coordinate system is the geometric center of the carrier device plane, the x-axis is parallel to the short side of the carrier device plane, and the y-axis is parallel to the long side of the carrier device plane.

[0089] Please see Figure 3 The top view image layer of the component is translated to the plane of the transport device to keep the reference point of the vertical load component coincide with the origin of the coordinate system of the transport plane;

[0090] In the plane of the carrier device, the edge contour of the sample vertical load component is divided into several edge pixels, and the edge pixels that coincide with the x-axis are marked as the initial edge pixels;

[0091] The edge contour of the transport device in the top view image of the component is marked to obtain the marked area of ​​the transport device. The origin of the coordinate system and the initial edge pixel are obtained to obtain the initial point connection line. The positive long axis of the vertical component is aligned with the initial point connection line. If the marked area of ​​the transport device can cover the image area of ​​the vertical load component, the initial edge pixel is divided into effective loading pixel. If the marked area of ​​the transport device cannot cover the image area of ​​the vertical load component, the initial edge pixel is divided into invalid loading pixel.

[0092] Repeat the process of classifying the initial edge pixels, classify each edge pixel separately, and when the edge point of the positive long axis of the vertical component coincides with the effective loading pixel, calculate the ratio of the area value of the vertical component image region to the area value of the marked region of the transport device to obtain the loading plane space ratio corresponding to the effective loading pixel, and compare the values ​​of the multiple loading plane space ratios obtained, and set the loading plane space ratio with the smallest value as the loading space coverage corresponding to the sample vertical component;

[0093] The effective loading pixel corresponding to the minimum loading plane space ratio is set as the loading feature pixel. The loading feature pixel is connected to the origin of the coordinate system to set the loading feature mark line. The angle between the loading feature mark line and the positive half axis of the x-axis at the origin of the coordinate system is obtained to obtain the preset loading angle of the component.

[0094] Repeat the process of obtaining the loading space coverage corresponding to the sample vertical load component, obtain the loading space coverage corresponding to each vertical load adapter component, and compare the values ​​of the obtained multiple loading space coverages. Set the vertical load adapter component corresponding to the maximum loading space coverage as the target adapter component.

[0095] The preset loading angle of the component corresponding to the target adapter component is obtained. The target adapter component is loaded using an AGV vehicle. During the loading process, the reference point of the vertical component is kept coincident with the geometric center point of the transport device plane. The angle between the positive major axis of the vertical component and the positive half axis of the x-axis of the transport plane coordinate system is the preset loading angle of the component. The AGV vehicle that has completed the loading of the target adapter component is set as the component loading vehicle.

[0096] The path analysis module sorts the transport routes of the component loading vehicles by distance, and performs straight-line trafficability analysis and curve trafficability analysis on the candidate routes in turn according to the sorting order, and obtains the target transport route based on the analysis results.

[0097] Specifically as follows:

[0098] Obtain the component loading vehicle, collect the regional map of the transportation area where the component loading vehicle is located, obtain the vehicle transportation area map, mark the current position of the component loading vehicle as the vehicle initial position on the vehicle transportation area map, mark the designated storage location of the target adapter component as the vehicle target position, obtain the passage path between the vehicle initial position and the vehicle target position, obtain multiple AVG vehicle passage paths, and mark the obtained AVG vehicle passage paths as candidate passage paths T1 to candidate passage paths Ta according to the path length from shortest to longest;

[0099] It should be noted here that:

[0100] In this application, the width of the AVG vehicle passage path mentioned herein is greater than the passage width required by the AVG vehicle;

[0101] In this application, candidate paths T1 to Ta are automatically generated by existing navigation terminals. In candidate paths T1 to Ta, T1 to Ta are the marker symbols corresponding to the AVG vehicle paths, and a is the number of AVG vehicle paths.

[0102] Perform passability analysis on candidate T1 routes, and identify abnormal passage areas based on the analysis results;

[0103] Specifically as follows:

[0104] The candidate T1 route is divided into straight sub-paths and curved sub-paths, and real-time planar image acquisition is performed on each sub-path to obtain multiple sub-path planar images.

[0105] It should be noted here that:

[0106] In this application, the method for dividing straight sub-paths and curved sub-paths is specifically to use curvature analysis to identify the curvature of candidate T1 paths, segmenting path regions with curvature greater than 0.05 rad / m into curved sub-paths, and segmenting path regions with curvature less than or equal to 0.05 rad / m into straight sub-paths.

[0107] Based on the sub-path planar image, the vehicle passability of the straight sub-path is analyzed for component loading, and the region type of the straight sub-path is divided according to the analysis results;

[0108] Specifically as follows:

[0109] In the sub-path planar image, the two sides of the straight sub-path are marked respectively to obtain the side edge of the first straight path and the side edge of the second straight path.

[0110] Obstacle recognition is performed based on the sub-path planar image. If there is an obstacle in the sub-path planar image, the obstacle outline is marked on the sub-path planar image to obtain the obstacle marking area. Edge pixels are extracted from the obstacle marking area. Perpendicular lines to the side of the first straight path are drawn through the edge pixels. The intersection of the drawn perpendicular lines and the side of the first straight path is marked as the intersection of the first side perpendicular lines. A straight line perpendicular to the side of the second straight path is drawn through the intersection of the first side perpendicular lines to obtain multiple obstacle-covering perpendicular lines.

[0111] The number of intersections between the edge of the obstacle marking area and the obstacle coverage vertical line is counted to obtain the number of obstacle coverage intersections. If the number of obstacle coverage intersections is equal to 1, the coverage length of the obstacle marking area to the obstacle coverage vertical line is numerically collected, and the difference between the obstacle coverage vertical line length value and the coverage length value is calculated to obtain the remaining passage distance corresponding to the obstacle coverage vertical line.

[0112] If the number of intersection points covered by obstacles is greater than 1, the intersection points of the obstacle marking area and the obstacle covering vertical line are collected to obtain multiple obstacle feature intersection points. The obstacle covering vertical line is divided into multiple obstacle sub-vertical lines using the obstacle feature intersection points. The length values ​​of the obstacle sub-vertical lines not covered by the obstacle marking area are collected, and the minimum length value is set as the remaining passage distance corresponding to the obstacle covering vertical line.

[0113] It should be noted here that:

[0114] If there are no obstacles in the sub-path planar image, then the minimum original width of the sub-path is set to the path passage width corresponding to the straight sub-path;

[0115] In this application, the obstacle referred to herein is specifically an object that is not movable for a short period of time, such as an AVG vehicle undergoing loading, unloading or repair operations.

[0116] The remaining passage distances corresponding to the vertical lines covered by different obstacles are compared numerically, and the remaining passage distance with the smallest value is marked as the path passage limit corresponding to the straight sub-path.

[0117] The required passage width of the component loading vehicle is numerically collected to obtain the required passage width of the vehicle. If the passage limit of the straight sub-path is less than or equal to the required passage width of the vehicle, the corresponding straight sub-area is marked as an abnormal passage area. If the passage limit of the straight sub-path is greater than the required passage width of the vehicle, the corresponding straight sub-area is marked as a normal passage area.

[0118] Based on the sub-path planar image, the vehicle passability analysis of the curved sub-path is performed, and the region type of the curved sub-path is divided according to the analysis results;

[0119] Specifically as follows:

[0120] In the sub-path planar image, the edges of the straight and curved paths on both sides are marked respectively to obtain the side of the first curved path and the side of the second curved path.

[0121] If there are no obstacles in the curved sub-path in the sub-path planar image, then the curved sub-path is analyzed for the curved radius to obtain the curved passage limit;

[0122] Specifically as follows:

[0123] In the sub-path planar image, the side of the first curve path is discretized into several side pixel points. Tangents are drawn to the side of the first curve path using the side pixel points as tangent points to obtain the side pixel tangents. Perpendiculars are drawn to the side pixel tangents through the side pixel points to obtain the side tangent point perpendicular lines. The intersection of the side tangent point perpendicular lines and the side of the first curve path is marked as the side perpendicular intersection pixel point. The midpoint of the line connecting each side pixel point and the corresponding side perpendicular intersection pixel point is collected to obtain multiple side connection midpoints. The side connection midpoints are connected sequentially by curves to obtain the sub-path curve feature curve.

[0124] Mark the start and end points of the curves in the sub-path curve feature curves respectively. Connect the start and end points of the curves to obtain the curve path feature line. Collect the length values ​​of the curve path feature straight lines to obtain the starting straight line distance of the curve. Obtain the midpoint of the sub-path curve feature curve to obtain the center point of the curve curve. Draw a perpendicular line from the center point of the curve curve to the curve path feature line to obtain the vertical distance of the curve high point.

[0125] The sub-path curve radius is obtained by calculating the starting straight distance of the curve and the vertical distance of the curve's highest point.

[0126] The curve radius of the sub-path is calculated using the following formula:

[0127] ;

[0128] Where Rwd is the curve radius of the sub-path, Hwd is the vertical distance from the curve's highest point, and Lwd is the straight-line distance from the curve's starting point.

[0129] It should be noted here that:

[0130] In this application, all curved sub-paths are approximated as circular arcs. In actual operation, the following test data exists:

[0131] 1 4 meters 8 meters 4 meters 2 5 meters 10 meters 5 meters 3 6 meters 12 meters 6 meters

[0132] If there are obstacles in the curved sub-path in the sub-path planar image, then the curve radius of the curved sub-path is analyzed to obtain the curve radius of the sub-path.

[0133] Specifically as follows:

[0134] In the sub-path planar image, obstacles are marked to obtain the path obstacle region. The edge of the path obstacle region is discretized into several obstacle edge pixels. The minimum line distance between each obstacle edge pixel and the side of the curve is obtained to obtain multiple obstacle curve side distances. The obstacle edge pixel corresponding to the maximum obstacle curve side distance is marked as the feature edge pixel, and the side of the curve corresponding to the maximum obstacle curve side distance is marked as the feature curve side.

[0135] It should be noted here that:

[0136] In this application, if the maximum obstacle bend side distance is the straight-line distance between the feature edge pixel and the side of the first bend path, then the side of the first bend path is the side of the feature bend.

[0137] The side of the feature curve is discretized into several feature side pixels. Then, a sample side pixel is randomly selected from the obtained side pixels. The sample side pixel is used as the tangent point to draw the tangent line of the feature curve side. The perpendicular line from the sample side pixel to the obtained tangent line is drawn to obtain the sample pixel tangent line. In the sample pixel tangent line, the pixels whose distance from the sample side pixel is the maximum obstacle curve side distance are marked to obtain the obstacle offset pixel corresponding to the sample side pixel.

[0138] Obtain the obstacle offset pixel corresponding to each feature side pixel, and connect the obstacle offset pixels in adjacent states using a smooth curve to obtain the offset curve side corresponding to the feature curve side.

[0139] The curve radius of the sub-path curve is obtained by taking the curve side of the feature curve and the curve side of the offset curve.

[0140] The required turning radius of the component loading vehicle is numerically collected to obtain the required turning radius of the vehicle. If the turning limit corresponding to the curve sub-path is less than or equal to the required turning radius of the vehicle, the corresponding straight sub-area is marked as an abnormal passage area. If the turning limit corresponding to the straight sub-path is greater than the required turning radius of the vehicle, the corresponding straight sub-area is marked as a normal passage area.

[0141] If there is no abnormal passage area in candidate T1, then candidate T1 will be selected as the target passage path. If there is an abnormal passage area in candidate T1, then the passageability analysis will be performed on candidate T2, and so on, until the target passage path is selected.

[0142] It should be noted here that:

[0143] In this application, among candidate routes T1 to Ta, at least one candidate route does not have an abnormal passage area. If no such route exists, a transportation anomaly warning is issued.

[0144] The delivery confirmation module intelligently transports the component loading vehicle according to the target travel path and performs accurate delivery confirmation of the target adapter component after transportation.

[0145] Specifically as follows:

[0146] The target passage path is obtained, the component loading vehicle transports the target adapter component according to the target passage path, unloads the target adapter component after transportation, and performs real-time planar image acquisition on the unloading area to obtain real-time monitoring image of unloading.

[0147] In the real-time monitoring image of unloading, the edge of the unloading location corresponding to the target adapter component is extracted to obtain the target unloading area, and the edge of the area where the target adapter component is currently located is extracted to obtain the real-time component location area.

[0148] If the real-time component location area is completely within the target unloading area, the accurate delivery of the target adapter component is completed. If the real-time component location area is not completely within the target unloading area, the unloading position of the target adapter component is adjusted until the real-time component location area is completely within the target unloading area.

[0149] Compared to the problems described in the background art, the present invention obtains the vertical loading height of large components by collecting and analyzing the loading height of the components, analyzes the vertical loading height of the components against the vertical loading limit of the carrier, selects vertical load-adaptive components based on the analysis results, analyzes the loading space coverage of the vertical load-adaptive components with the AGV to be loaded, determines the target adaptable components based on the analysis results and analyzes the loading angle to obtain the preset loading angle of the components, and intelligently loads the target adaptable components according to the preset loading angle of the components. This can make full use of the transportation space and further improve the transportation efficiency of AGV transportation equipment.

[0150] Furthermore, this invention sorts the distance of the transport path for the component loading vehicle, and performs straight-line trafficability analysis and curve trafficability analysis on the candidate transport paths in turn according to the sorting order. Based on the analysis results, the target transport path is obtained, which can effectively screen out the shortest and most trafficable path, thereby improving transportation efficiency and safety.

[0151] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. An AGV intelligent loading and precise delivery system for large components to designated storage locations, characterized in that, include: Component loading module: Collects and analyzes the loading height of large components to obtain vertical load adaptors. Performs loading space coverage analysis and loading angle analysis between the vertical load adaptors and the AGV to be loaded to obtain the preset loading angle of the components. Intelligently loads the target adaptors according to the preset loading angle of the components to obtain the component loading vehicle. Path analysis module: Sort the distance of the transport path for the component loading vehicle, and perform straight road trafficability analysis and curve trafficability analysis on the candidate transport paths in turn according to the sorting order, and obtain the target transport path based on the analysis results; Delivery confirmation module: Intelligently transports the component loading vehicle according to the target travel path, and performs accurate delivery confirmation of the target adapter component after transportation.

2. The AGV intelligent loading and precise delivery system for large components to designated storage locations according to claim 1, characterized in that, The component loading vehicle is acquired as follows: Obtain the AGV to be loaded, and obtain the maximum loading height corresponding to the AGV to be loaded to get the vertical loading limit of the vehicle; The maximum ground clearance of large components is obtained to obtain the vertical height of the components. The ground clearance of the AGV's loading device is obtained to obtain the ground clearance of the loading device. The sum of the vertical height of the components and the ground clearance of the loading device is calculated to obtain the vertical loading height of the components. Large components whose vertical loading height is less than the vertical loading limit of the vehicle are marked as vertical load adaptable components. Sample vertical load components are selected from the obtained vertical load adaptable components. The loading space coverage of the sample vertical load component and the AGV to be loaded is analyzed. Based on the analysis results, the loading space coverage and the preset loading angle of the component are obtained. Obtain the loading space coverage corresponding to each vertical load adapter component, compare the values ​​of the multiple obtained loading space coverage, and set the vertical load adapter component corresponding to the maximum loading space coverage as the target adapter component. The preset loading angle of the component corresponding to the target adapter component is obtained. The target adapter component is loaded using an AGV vehicle according to the preset loading angle. The AGV vehicle that has completed loading the target adapter component is set as the component loading vehicle.

3. The AGV intelligent loading and precise delivery system for large components to designated storage locations according to claim 2, characterized in that, The loading space coverage and the preset loading angle of the components are obtained as follows: Planar radiometric images of the sample vertical load component are acquired to obtain a top view image of the component. A vertical load component image region and a circumcircle of the vertical load component are created in the top view image of the component. The center of the circumcircle of the vertical load component is marked as the reference point of the vertical load component. The major axis of the sample vertical load component is identified. The identification result is used to obtain the positive major axis of the vertical component. The plane of the carrier device is obtained and a plane rectangular coordinate system is created to obtain the carrier plane coordinate system. The top view image layer of the component is translated to the plane of the transport device to keep the reference point of the vertical load component coincide with the origin of the coordinate system of the transport plane; In the plane of the carrier device, the edge contour of the sample vertical load component is divided into several edge pixels. The edge pixels are divided into effective edge pixels and ineffective edge pixels. When the edge point of the positive long axis of the vertical component coincides with the effective loading pixel, the ratio of the area value of the vertical load component image region to the area value of the carrier device marking region is calculated to obtain the loading plane space ratio corresponding to the effective loading pixel. The loading plane space ratio with the smallest value is set as the loading space coverage corresponding to the sample vertical load component. Obtain the loading feature pixels, connect the loading feature pixels to the origin of the coordinate system to set the loading feature mark line, and obtain the angle value between the loading feature mark line and the positive half axis of the x-axis at the origin of the coordinate system to obtain the preset loading angle of the component.

4. The AGV intelligent loading and precise delivery system for large components to designated storage locations according to claim 3, characterized in that, The positive major axis of the vertical component is obtained as follows: A circumferential pixel window is set for the outer circle of the vertical load component, and a feature circumferential pixel point is arbitrarily selected on the outer circle of the vertical load component. When the circumferential pixel window is at the feature circumferential pixel point, the intercept distance of the component diameter corresponding to the feature circumferential pixel point is obtained. Using the feature circumferential pixel point as the starting point, the system reverses the circumference of half of the vertical component by traversing the circumference of the circumferential pixel window. The system obtains the component diameter intercept distance corresponding to different circumferential pixel points and sets the component diameter intercept distance with the largest value as the major axis distance of the vertical component. The first diameter edge intersection point and the second diameter edge intersection point corresponding to the major axis distance of the vertical load component are obtained. The distance value between the first diameter edge intersection point and the reference point of the vertical load component is obtained to obtain the first intersection point intercept distance. The distance value between the second diameter edge intersection point and the reference point of the vertical load component is obtained to obtain the second intersection point intercept distance. If the first intersection point intercept distance is greater than or equal to the second intersection point intercept distance, the line connecting the first diameter edge intersection point and the reference point of the vertical load component is set as the positive major axis of the vertical load component. If it is less than, the line connecting the second diameter edge intersection point and the reference point of the vertical load component is set as the positive major axis of the vertical load component.

5. The AGV intelligent loading and precise delivery system for large components to designated storage locations according to claim 1, characterized in that, The target travel path is obtained as follows: Obtain the component loading vehicle, collect the regional map of the transportation area where the component loading vehicle is located, obtain the vehicle transportation area map, mark the current position of the component loading vehicle as the vehicle initial position on the vehicle transportation area map, mark the specified storage location of the target adapter component as the vehicle target position, obtain the passage path between the vehicle initial position and the vehicle target position, obtain multiple AVG vehicle passage paths, and mark them as candidate passage paths T1 to Ta according to the path length from shortest to longest; Perform passability analysis on candidate T1 routes, and identify abnormal passage areas based on the analysis results; If there is no abnormal passage area in candidate path T1, then candidate path T1 is selected as the target passage path. If there is an abnormal passage area in candidate path T1, then the passageability analysis is performed on candidate path T2, and so on, until the target passage path is selected.

6. The AGV intelligent loading and precise delivery system for large components to designated storage locations according to claim 5, characterized in that, The abnormal passage area is identified for candidate passage path T1, as follows: The candidate T1 route is divided into straight sub-paths and curved sub-paths. Real-time planar image acquisition is performed on each sub-path to obtain multiple sub-path planar images. Based on the sub-path planar images, the vehicle passability analysis of straight and curved sub-paths is performed to identify abnormal passage areas. The straight sub-paths are divided into region types as follows: In the sub-path planar image, the two sides of the straight sub-path are marked respectively to obtain the side edge of the first straight path and the side edge of the second straight path. Obstacle recognition is performed based on the sub-path planar image. If there is an obstacle in the sub-path planar image, the obstacle outline is marked on the sub-path planar image to obtain the obstacle marking area. Edge pixels are extracted from the obstacle marking area. A perpendicular line is drawn from the edge pixel to the side of the first straight path. The intersection of the perpendicular line and the side of the first straight path is marked as the intersection point of the first side perpendicular line. A straight line perpendicular to the side of the second straight path is drawn from the intersection point of the first side perpendicular line to obtain multiple obstacle-covering perpendicular lines.

7. The AGV intelligent loading and precise delivery system for large components to designated storage locations according to claim 6, characterized in that, The straight sub-paths are divided into region types as follows: The number of intersections between the edge of the obstacle marking area and the obstacle coverage vertical line is counted to obtain the number of obstacle coverage intersections. If the number of obstacle coverage intersections is equal to 1, the coverage length of the obstacle marking area to the obstacle coverage vertical line is numerically collected, and the difference between the obstacle coverage vertical line length value and the coverage length value is calculated to obtain the remaining passage distance corresponding to the obstacle coverage vertical line. If the number of intersection points covered by obstacles is greater than 1, the intersection points of the obstacle marking area and the obstacle covering vertical line are collected to obtain multiple obstacle feature intersection points. The obstacle covering vertical line is divided into multiple obstacle sub-vertical lines using the obstacle feature intersection points. The length values ​​of the obstacle sub-vertical lines not covered by the obstacle marking area are collected, and the minimum length value is set as the remaining passage distance corresponding to the obstacle covering vertical line. The remaining passage distances corresponding to the vertical lines covered by different obstacles are compared numerically, and the remaining passage distance with the smallest value is marked as the path passage limit corresponding to the straight sub-path. The required passage width of the component loading vehicle is numerically collected to obtain the required passage width of the vehicle. If the passage limit of the straight sub-path is less than or equal to the required passage width of the vehicle, the corresponding straight sub-area is marked as an abnormal passage area. If the passage limit of the straight sub-path is greater than the required passage width of the vehicle, the corresponding straight sub-area is marked as a normal passage area.

8. The AGV intelligent loading and precise delivery system for large components to designated storage locations according to claim 7, characterized in that, The curve sub-paths are categorized by region type as follows: In the sub-path planar image, the edges of the straight and curved paths on both sides are marked respectively to obtain the side of the first curved path and the side of the second curved path. If there are no obstacles in the curved sub-path in the sub-path planar image, then the curve radius of the curved sub-path is analyzed to obtain the curve radius of the sub-path. If there are obstacles in the curved sub-path in the sub-path planar image, then the curve radius of the curved sub-path is analyzed to obtain the curve radius of the sub-path. The required turning radius of the component loading vehicle is numerically collected to obtain the required turning radius of the vehicle. If the turning limit corresponding to the curve sub-path is less than or equal to the required turning radius of the vehicle, the corresponding straight sub-area is marked as an abnormal passage area. If the turning limit corresponding to the straight sub-path is greater than the required turning radius of the vehicle, the corresponding straight sub-area is marked as a normal passage area.

9. The AGV intelligent loading and precise delivery system for large components to designated storage locations according to claim 8, characterized in that, The curve radius of the sub-path is obtained as follows: In the sub-path planar image, the side of the first curve path is discretized into several side pixel points. Tangents are drawn to the side of the first curve path using the side pixel points as tangent points to obtain the side pixel tangents. Perpendiculars are drawn to the side pixel tangents through the side pixel points to obtain the side tangent point perpendicular lines. The intersection of the side tangent point perpendicular lines and the side of the first curve path is marked as the side perpendicular intersection pixel point. The midpoint of the line connecting each side pixel point and the corresponding side perpendicular intersection pixel point is collected to obtain multiple side connection midpoints. The side connection midpoints are connected sequentially by curves to obtain the sub-path curve feature curve. Mark the start and end points of the curves in the sub-path curve feature curves respectively. Connect the start and end points of the curves to obtain the curve path feature line. Collect the length values ​​of the curve path feature straight lines to obtain the starting straight line distance of the curve. Obtain the midpoint of the sub-path curve feature curve to obtain the center point of the curve curve. Draw a perpendicular line from the center point of the curve curve to the curve path feature line to obtain the vertical distance of the curve high point. The sub-path curve radius is obtained by calculating the starting straight distance of the curve and the vertical distance from the highest point of the curve.

10. The AGV intelligent loading and precise delivery system for large components to designated storage locations according to claim 8, characterized in that, The curve radius of the sub-path is obtained as follows: In the sub-path planar image, obstacles are marked to obtain the path obstacle region. The edge of the path obstacle region is discretized into several obstacle edge pixels. The minimum line distance between the obstacle edge pixels and the side of the curve is obtained to obtain multiple obstacle curve side distances. The obstacle edge pixels corresponding to the maximum obstacle curve side distance are marked as feature edge pixels, and the side of the curve corresponding to the maximum obstacle curve side distance is marked as feature curve side. The side of the feature curve is discretized into several feature side pixels. A sample side pixel is arbitrarily selected. The sample side pixel is used as the tangent point to draw the tangent line of the feature curve side. A perpendicular line is drawn through the sample side pixel to the obtained tangent line to obtain the sample pixel tangent line. In the sample pixel tangent line, the pixels that are at the maximum obstacle curve side distance from the sample side pixel are marked to obtain the obstacle offset pixel corresponding to the sample side pixel. Obtain the obstacle offset pixel corresponding to each feature side pixel, and use a smooth curve to connect the obstacle offset pixels in adjacent states to obtain the offset curve side corresponding to the feature curve side. The curve radius of the sub-path curve is obtained by taking the curve side of the feature curve and the curve side of the offset curve.

Images