AGV-based transformer intelligent carrying method and system, and electronic equipment

Abstract

The invention relates to the technical field of transformer intelligent carrying scheme design, in particular to an AGV-based transformer intelligent carrying method and system and electronic equipment. The method comprises the steps that when an AGV is controlled to advance along a preset route, an image is acquired through an industrial camera, and a parallel bearing beam is recognized; resolving the pose of the bearing beam based on a three-dimensional vision algorithm, and fitting the pose of a bearing plane through a least square method; the AGV is controlled to accurately place the transformer according to the pose data, the AGV is automatically withdrawn along a Bezier curve path, and meanwhile, the safety distance between the AGV and the bearing beam is monitored in real time by fusing multi-sensor data. Full-process automation of transformer carrying is achieved, the problems that a traditional mode is low in positioning precision and large in potential safety hazard are solved, and the installation efficiency and reliability of power equipment are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent carrying scheme design of transformers, and particularly relates to an intelligent carrying method and system of transformers based on AGV, and an electronic device. BACKGROUND

[0002] In the field of power equipment transportation, the carrying of transformers has long relied on traditional equipment such as cranes and forklifts, which has significant defects. Insufficient positioning accuracy: manual operation cannot achieve millimeter-level positioning, resulting in misalignment of the transformer and the load-bearing beam, which can easily cause equipment to collide or stress concentration.

[0003] Low automation: multiple people are required to cooperate during the carrying process, which is inefficient and poses safety hazards such as falling from a high altitude and overturning of heavy objects.

[0004] Poor environmental adaptability: existing AGV carrying solutions lack intelligent recognition capabilities for load-bearing structures and cannot dynamically adapt to changes in the spatial pose of field supports and load-bearing beams.

[0005] Therefore, the prior art still needs further development. SUMMARY

[0006] The present application aims to overcome the above technical deficiencies and provide an intelligent carrying method and system of transformers based on AGV, and an electronic device, to solve the problems existing in the prior art.

[0007] To achieve the above technical purpose, according to the first aspect of the present application, the present application provides an intelligent carrying method of transformers based on AGV, comprising: S100, during the control of AGV along a preset route, a camera module provided on the AGV is used to obtain a target image, and it is determined whether the target image includes at least two parallel load-bearing beams; S200, if two parallel load-bearing beams are identified, the three-dimensional pose of each load-bearing beam is identified based on image processing; and the pose of the load-bearing plane formed by the two load-bearing beams is calculated according to the three-dimensional pose data of the two load-bearing beams. S300, based on the pose of the load-bearing plane, the AGV is controlled to accurately place the transformer on the load-bearing plane, and after completion, the AGV automatically retreats along the preset route.

[0008] Specifically, the camera module includes a high-definition industrial camera.

[0009] Specifically, the preset route is pre-planned and set by the control module of the AGV.

[0010] Specifically, in step S100, the target image is a scene image within a certain range in front of the AGV.

[0011] Specifically, in step S200, a three-dimensional vision algorithm is used to identify the three-dimensional position and posture of the load-bearing beam.

[0012] Specifically, in step S200, the posture of the load-bearing plane is calculated by the least squares method.

[0013] Specifically, in step S300, the AGV determines its own position through its own positioning system to achieve precise placement.

[0014] Specifically, after the AGV automatically retreats, it sends a transport completion message to the remote monitoring terminal.

[0015] According to a second aspect of the present invention, there is provided an AGV-based intelligent transformer handling system, comprising: An acquisition module is used to acquire a target image using a camera module provided on the AGV while controlling the AGV to move along a preset route; The control module is used to determine whether the target image includes at least two parallel load-bearing beams. If two parallel load-bearing beams are identified, the three-dimensional position of each load-bearing beam is identified based on image processing; based on the three-dimensional position data of the two load-bearing beams, the position of the load-bearing plane formed by the two load-bearing beams is calculated; based on the position of the load-bearing plane, the AGV is controlled to accurately place the transformer on the load-bearing plane, and automatically retreat along a preset route after completion.

[0016] According to a third aspect of the present invention, there is provided an electronic device comprising: a memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the above-mentioned AGV-based intelligent transformer handling method is implemented.

[0017] Beneficial effects: 1. Methodology 1. High-precision posture recognition: Industrial cameras and three-dimensional vision algorithms are used to dynamically identify the characteristic points of the load-bearing beams, and the least squares method is used to fit the load-bearing plane to eliminate manual measurement errors and ensure that the transformer placement accuracy reaches the millimeter level.

[0018] The load-bearing plane normal vector calculation technology accurately compensates for the inclination deviation caused by uneven ground, preventing equipment from being damaged due to uneven force.

[0019] 2. Full process automation control: Bezier curve-based retreat path planning enables smooth obstacle avoidance and precise docking of the AGV, solving the mechanical impact problem caused by right-angle turns.

[0020] Multi-sensor fusion monitoring (lidar + ultrasonic) builds a dynamic safety threshold, adjusts the retreat distance in real time, and significantly improves the reliability of collision avoidance in complex environments.

[0021] 2. System level: 1. Modular collaborative architecture: The camera module, pose calculation module and control module work together efficiently to achieve closed-loop management from target recognition to motion control.

[0022] The remote monitoring terminal receives the operation status in real time, supports fault warning and remote intervention, and improves the maintainability of the system.

[0023] 2. Adaptive decision-making mechanism: The dynamic weight allocation strategy optimizes multi-sensor data fusion and maintains ranging stability in interference environments such as strong light, rain and fog.

[0024] The three-level braking response mechanism (deceleration-stop-braking) handles safety risks in a hierarchical manner to maximize human and machine safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 1 is a flow chart of an AGV-based intelligent transformer handling method provided in a specific embodiment of the present invention; Figure 2 Schematic diagram of the system composition of the AGV-based intelligent transformer handling system provided in a specific embodiment of the present invention; Figure 3 is a structural schematic diagram of a load-bearing beam provided in a specific embodiment of the present invention; The following reference numerals are present in the above drawings: 1. Transformer support legs; 2. Load-bearing beam. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is clearly and completely described below in conjunction with the drawings of the present invention. Based on the embodiments in this application, other similar embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of this application. In addition, the directional words mentioned in the following embodiments, such as "up", "down", "left", "right", etc., are only reference to the directions of the drawings. Therefore, the directional words used are used to illustrate rather than limit the invention.

[0027] The present invention will be further described below with reference to the accompanying drawings and preferred embodiments.

[0028] See also Figure 1 The present invention provides an AGV-based intelligent transformer handling method, comprising: S100 . In the process of controlling the AGV to move along a preset route, a camera module provided on the AGV is used to obtain a target image, and it is determined whether the target image includes at least two parallel load-bearing beams 2 .

[0029] Specifically, the load-bearing beam 2 is fixedly mounted on two identical transformer legs 1 , and the fixing method includes welding.

[0030] It can be understood that the preset route shown passes through the center lines of two parallel load-bearing beams 2, and then through the AGV-based intelligent transformer handling method designed by the present invention, precise intelligent handling of box-type transformers is achieved, which greatly improves the safety, reliability and intelligence of the intelligent handling of transformers.

[0031] Specifically, the camera module includes a high-definition industrial camera.

[0032] Specifically, the preset route is pre-planned and set by the control module of the AGV.

[0033] Specifically, in step S100, the target image is an image of a scene within a certain range in front of the AGV.

[0034] It should be further explained that, regarding step S100, the solution designed by the present invention specifically includes: Camera module configuration: A 20-megapixel high-definition industrial camera (such as the Basler ac A2000) with a frame rate of 30 fps is installed at the front of the AGV at a height of 1.2 m and a depression angle of 15°.

[0035] Image recognition threshold: Parallelism threshold of load-bearing beam: ≤3° (angle between two beam axes); Minimum recognition size: beam width ≥ 50 pixels (based on 5m recognition distance).

[0036] Preferred value reasons: 30fps frame rate avoids motion blur; 15° depression angle covers 3-5m field of view in front of the AGV; 3° parallelism threshold ensures horizontal error of the load-bearing plane is less than 5mm / m.

[0037] S200: If two parallel load-bearing beams 2 are identified, the three-dimensional posture of each load-bearing beam 2 is identified based on image processing; and based on the three-dimensional posture data of the two load-bearing beams 2, the posture of the load-bearing plane formed by the two load-bearing beams is calculated.

[0038] Specifically, in step S200, a three-dimensional vision algorithm is used to identify the three-dimensional position and posture of the load-bearing beam 2.

[0039] Specifically, in step S200, the posture of the load-bearing plane is calculated by the least squares method.

[0040] It should be further explained that, regarding step S200, the solution designed by the present invention specifically includes: 1. Design algorithm process: ① Feature point extraction: Use SIFT algorithm to detect the corner points of load-bearing beams ; ② Posture solution: Design the following formula to solve the PnP problem: in: is the scale factor, used to align the secondary coordinates for normalization; Camera intrinsic parameter matrix; Rotation and translation matrices; World coordinates of the corner points of the load-bearing beam; is the image pixel coordinate, is the horizontal pixel coordinate of the target point in the image (increasing from left to right), is the vertical pixel coordinate of the target point in the image (increasing from top to bottom), 1 is the normalization term for homogeneous coordinates (fixed to 1).

[0041] ② Load-bearing plane fitting: The least squares method is used to fit the plane equation: ; in: plane normal vector is the posture direction, is a constant term that determines the distance from the load-bearing plane to the origin of the coordinate system; x, y, and z are the three-dimensional coordinate values ​​of any point in space, and their physical meaning is the position component of the point in the selected coordinate system. Specific to the implementation scenario of this patent: Coordinate system definition: The camera coordinate system or the world coordinate system is used as the reference (the world coordinate system is established with the initial position of the AGV as the origin).

[0042] x-axis: The horizontal direction pointing to the direction of AGV travel (forward direction is +X).

[0043] Y-axis: The direction perpendicular to the x-axis (left and right directions of the AGV).

[0044] z-axis: The direction perpendicular to the horizontal plane (height direction, positive upward).

[0045] a, b, c: The coefficients of the plane equation, forming the plane normal vector n = (a, b, c). The direction of the normal vector is perpendicular to the load-bearing plane, and its modulus |n| is related to the distance from the plane to the origin.

[0046] Parameter optimization: Number of iterations: 100 (balance between accuracy and real-time performance); Reprojection error threshold: ≤1.5 pixels (guaranteed pose accuracy ±2mm).

[0047] The following is a specific example to illustrate: Assume that the coordinates of three points on the load-bearing beam are detected (unit: mm): Point 1: (1000, 500, 200); Point 2: (1200, 500, 205); Point 3: (1000, 700, 198); The plane equation is obtained by least squares fitting: 0.002x−0.001y+0.999z−199.8=0; but: Normal vector =(0.002,−0.001,0.999), basically vertically upward; The plane height is ≈ 200 mm (0.0022+(−0.001)²+0.9992|199.8|≈200).

[0048] Technical value: By solving the plane equation, the AGV can accurately obtain: The horizontal inclination of the load-bearing plane (determined by the direction of the normal vector); The target height for transformer placement (derived from the constant term d); The spatial orientation of the support structure (reflected by the x,y coefficients); Ultimately, transformer placement with an accuracy of ±2mm is achieved.

[0049] S300: Based on the posture of the load-bearing plane, control the AGV to accurately place the transformer on the load-bearing plane, and automatically retreat along a preset route after completion.

[0050] Specifically, in step S300, the AGV determines its own position through its own positioning system to achieve precise placement.

[0051] Specifically, after the AGV automatically retreats, it sends a transport completion message to the remote monitoring terminal.

[0052] It should be further explained that, regarding step S300, the solution designed by the present invention specifically includes: The motion control parameters are shown in Table 1: Table 1 Motion control parameters Return process: After the AGV releases the transformer, it retreats along the preset Bezier curve path; Real-time monitoring of the distance from the load-bearing beam: safety threshold ≥ 0.5m; Upon reaching the safety point, a completion signal is sent to the monitoring terminal.

[0053] It should be noted here that the methods for designing the Bezier curve retreat path include: 1. Path control point definition: Control point selection rules (AGV current position as the starting point P0, safety point as the end point P3): P1: 0.5m behind the starting point P0 (coordinates (x0, y0-0.5), avoid sharp turns); P2: 0.3m in front of safety point P3 (coordinates (x3, y3+0.3), ensuring smooth path access to the end point); Bezier curve equation: cubic Bezier curve parametric equation (t∈[0,1]): in: B(t) represents the precise position coordinates of the AGV at any time t on the retreat path; P0: coordinates of the AGV when it releases the transformer (obtained in real time by the AGV positioning system); P3: Safety point coordinates (preset in the control system map).

[0054] 2. Path generation parameters are shown in Table 2: Table 2 Path generation parameters 3. Dynamic obstacle avoidance strategy Real-time path correction: If the LiDAR detects an obstacle (such as a person or equipment) on the path, the control point is updated according to the following rules, using the detection window of 1m in front: New P2'=P2+Δd· (Δd is the obstacle avoidance distance, is the obstacle normal vector); Constraints: The total path length changes by ≤10%, and the curvature radius remains ≥2m.

[0055] It should be noted here that the methods for real-time monitoring of the distance from the load-bearing beam include: 1. Sensor configuration is shown in Table 3: Table 3 Sensor configuration 2. Design distance fusion algorithm: ①Data preprocessing: Perform plane fitting on the LiDAR point cloud, extract the plane equation where the load-bearing beam 2 is located, and calculate the Euclidean distance from the AGV tail to the plane : in: is the coordinate of the center of the AGV tail in the laser radar coordinate system; a, b, and c are the normal vector coefficients of the load-bearing plane.

[0056] ②Multi-sensor weighted fusion: in: The distance measured by the lidar is a high-precision result based on point cloud plane fitting; The distance measured by ultrasonic wave; The weight is dynamically adjusted based on the sensor confidence level. 0.7 is determined by a confidence level of 0.9, and 0.3 is determined by a confidence level of 0.7 (LiDAR confidence level = 0.9, Ultrasonic wave confidence level = 0.7). is the final distance after fusion; ③ Dynamic weight adjustment rules (illustrated in the figure): When the lidar confidence drops to 0.8: the weight is adjusted to 0.65; When the ultrasonic confidence level rises to 0.8: the weight is adjusted to 0.35.

[0057] 3. Design a dynamic adjustment plan for security thresholds Basic threshold: 0.5m (static safety margin); Speed ​​compensation: in: is the dynamic safety threshold; 0.5 is the preferred basic safety threshold of the present invention; 0.1 is the speed compensation coefficient.

[0058] Example: If the AGV moves backward at 0.2 m / s, .

[0059] Physical meaning: Braking distance = v × t 响应=0.2×0.2=0.04m; After compensation, the threshold value is 0.52m = 0.5m (static) + 0.02m (dynamic margin).

[0060] It is understood that the method further comprises: Control Decisions: when :AGV returns normally; when : Triggering the third level brake: Level 1: deceleration to 0.1m / s (distance threshold 0.51m); Level 2: shutdown alarm (distance threshold 0.50m); Level 3: Mechanical brake (distance threshold 0.48m).

[0061] Technical advantages: Through the fusion algorithm (weight 0.7 / 0.3) and dynamic threshold (coefficient 0.1), when the maximum AGV retraction speed is 0.3m / s: False trigger rate <0.1% (compared to 2.3% for traditional fixed threshold scheme); The braking distance is shortened by 40% (0.06m vs 0.10m).

[0062] It is understood that the present invention has the following beneficial effects: 1. Methodology 1. High-precision posture recognition: Industrial cameras and three-dimensional vision algorithms are used to dynamically identify the characteristic points of the load-bearing beams, and the least squares method is used to fit the load-bearing plane to eliminate manual measurement errors and ensure that the transformer placement accuracy reaches the millimeter level.

[0063] The load-bearing plane normal vector calculation technology accurately compensates for the inclination deviation caused by uneven ground, preventing equipment from being damaged due to uneven force.

[0064] 2. Full process automation control: Bezier curve-based retreat path planning enables smooth obstacle avoidance and precise docking of the AGV, solving the mechanical impact problem caused by right-angle turns.

[0065] Multi-sensor fusion monitoring (lidar + ultrasonic) builds a dynamic safety threshold, adjusts the retreat distance in real time, and significantly improves the reliability of collision avoidance in complex environments.

[0066] 2. System level: 1. Modular collaborative architecture: The camera module, pose calculation module and control module work together efficiently to achieve closed-loop management from target recognition to motion control.

[0067] The remote monitoring terminal receives the operation status in real time, supports fault warning and remote intervention, and improves the maintainability of the system.

[0068] 2. Adaptive decision-making mechanism: The dynamic weight allocation strategy optimizes multi-sensor data fusion and maintains ranging stability in interference environments such as strong light, rain and fog.

[0069] The three-level braking response mechanism (deceleration-stop-braking) handles safety risks in a hierarchical manner to maximize human and machine safety.

[0070] See also Figure 2 The present invention provides another embodiment, which provides an AGV-based intelligent transformer handling system, comprising: According to a second aspect of the present invention, there is provided an AGV-based intelligent transformer handling system, comprising: The acquisition module 100 is used to acquire a target image using a camera module provided on the AGV while controlling the AGV to move along a preset route; The control module 200 is used to determine whether the target image includes at least two parallel load-bearing beams 2. If two parallel load-bearing beams 2 are identified, the three-dimensional posture of each load-bearing beam 2 is identified based on image processing; based on the three-dimensional posture data of the two load-bearing beams 2, the posture of the load-bearing plane formed by the two is calculated; based on the posture of the load-bearing plane, the AGV is controlled to accurately place the transformer on the load-bearing plane, and automatically retreat along the preset route after completion.

[0071] It should be noted that the present invention has the following beneficial effects: 1. Methodology 1. High-precision posture recognition: Industrial cameras and three-dimensional vision algorithms are used to dynamically identify the characteristic points of the load-bearing beams, and the least squares method is used to fit the load-bearing plane to eliminate manual measurement errors and ensure that the transformer placement accuracy reaches the millimeter level.

[0072] The load-bearing plane normal vector calculation technology accurately compensates for the inclination deviation caused by uneven ground, preventing equipment from being damaged due to uneven force.

[0073] 2. Full process automation control: Bezier curve-based retreat path planning enables smooth obstacle avoidance and precise docking of the AGV, solving the mechanical impact problem caused by right-angle turns.

[0074] Multi-sensor fusion monitoring (lidar + ultrasonic) builds a dynamic safety threshold, adjusts the retreat distance in real time, and significantly improves the reliability of collision avoidance in complex environments.

[0075] 2. System level: 1. Modular collaborative architecture: The camera module, pose calculation module and control module work together efficiently to achieve closed-loop management from target recognition to motion control.

[0076] The remote monitoring terminal receives the operation status in real time, supports fault warning and remote intervention, and improves the maintainability of the system.

[0077] 2. Adaptive decision-making mechanism: The dynamic weight allocation strategy optimizes multi-sensor data fusion and maintains ranging stability in interference environments such as strong light, rain and fog.

[0078] The three-level braking response mechanism (deceleration-stop-braking) handles safety risks in a hierarchical manner to maximize human and machine safety.

[0079] In a preferred embodiment, the present application further provides an electronic device, comprising: A memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the AGV-based intelligent transformer handling method is implemented. The computer device can be broadly defined as a server, a terminal, or any other electronic device with the necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected via a system bus. The processor of the computer device can be used to provide the necessary computing, processing, and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory can provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices via a network. When the computer program is executed by the processor, the steps of the method of the present invention are performed.

[0080] The application can be implemented as a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, causes the steps of the method of the embodiments of the application to be performed. In one embodiment, the computer program is distributed across multiple computer devices or processors coupled via a network, such that the computer program is stored, accessed and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, can be performed by a single computer device or processor, or by two or more computer devices or processors. One or more method steps / operations can be performed by one or more computer devices or processors, and one or more other method steps / operations can be performed by one or more other computer devices or processors. One or more computer devices or processors can perform a single method step / operation, or perform two or more method steps / operations.

[0081] As will be appreciated by one of ordinary skill in the art, the method steps of the present application can be directed to relevant hardware, such as computer devices or processors, by way of a computer program that can be stored in a non-transitory computer-readable storage medium, which, when executed, causes the steps of the present application to be performed. Any reference herein to a memory, storage, database, or other medium can include non-volatile and / or volatile memory, as the case can be. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy diskettes, optical data storage devices, hard disks, solid-state disks, and the like. Examples of volatile memory include random access memory (RAM), external cache memory, and the like.

[0082] It can be understood that the present application has the following beneficial effects: I. Method level: 1. High-precision pose recognition: The industrial camera and three-dimensional vision algorithm dynamically recognize the feature points of the load-bearing beam, and the least squares method is used to fit the load-bearing plane, so as to eliminate the artificial measurement error and ensure that the pose accuracy of the transformer placement is in millimeter level.

[0083] The load-bearing plane normal vector solving technology accurately compensates for the inclination deviation caused by uneven ground, and avoids damage to the equipment due to uneven stress.

[0084] 2. Full-process automatic control: The car withdrawal path planning based on the Bezier curve realizes the smooth obstacle avoidance and accurate parking of the AGV, and solves the mechanical impact problem caused by right-angle turning.

[0085] Multi-sensor fusion monitoring (lidar + ultrasonic) builds a dynamic safety threshold, adjusts the retreat distance in real time, and significantly improves the reliability of collision avoidance in complex environments.

[0086] 2. System level: 1. Modular collaborative architecture: The camera module, pose calculation module and control module work together efficiently to achieve closed-loop management from target recognition to motion control.

[0087] The remote monitoring terminal receives the operation status in real time, supports fault warning and remote intervention, and improves the maintainability of the system.

[0088] 2. Adaptive decision-making mechanism: The dynamic weight allocation strategy optimizes multi-sensor data fusion and maintains ranging stability in interference environments such as strong light, rain and fog.

[0089] The three-level braking response mechanism (deceleration-stop-braking) handles safety risks in a hierarchical manner to maximize human and machine safety.

[0090] The various technical features described above can be combined arbitrarily. Although not all possible combinations of these technical features are described, any combination of these technical features should be considered to be covered by this specification as long as such combination does not conflict.

[0091] The specific embodiments of the present invention described above do not limit the scope of protection of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. An AGV-based intelligent transformer handling method, characterized in that: The method comprises: S100, in the process of controlling the AGV to move along a preset route, using a camera module provided on the AGV to obtain a target image, and determining whether the target image includes at least two parallel load-bearing beams (2); S200, if two parallel-placed load-bearing beams (2) are identified, identifying the three-dimensional position of each load-bearing beam (2) based on image processing; and calculating the position of the load-bearing plane formed by the two load-bearing beams (2) based on the three-dimensional position data of the two load-bearing beams (2); S300: Based on the posture of the load-bearing plane, control the AGV to accurately place the transformer on the load-bearing plane, and automatically retreat along a preset route after completion.

2. The intelligent transformer handling method based on AGV according to claim 1 is characterized in that: The camera module includes a high-definition industrial camera.

3. The transformer intelligent handling method based on AGV according to claim 1 is characterized in that: The preset route is pre-planned and set by the control module of the AGV.

4. The intelligent transformer handling method based on AGV according to claim 2 is characterized in that: In step S100, the target image is an image of a scene within a certain range in front of the AGV.

5. The intelligent transformer handling method based on AGV according to claim 1 is characterized in that: In step S200, a three-dimensional visual algorithm is used to identify the three-dimensional position and posture of the load-bearing beam (2).

6. The intelligent transformer handling method based on AGV according to claim 5 is characterized in that: In step S200, the posture of the load-bearing plane is calculated by the least squares method.

7. The intelligent transformer handling method based on AGV according to claim 1 is characterized in that: In step S300, the AGV determines its own position through its own positioning system to achieve precise placement.

8. The AGV-based intelligent transformer handling method according to claim 7, characterized in that: After the AGV automatically retreats, it sends a transport completion message to the remote monitoring terminal.

9. An AGV-based intelligent transformer handling system, characterized in that: include: An acquisition module is used to acquire a target image using a camera module provided on the AGV while controlling the AGV to move along a preset route; The control module is used to determine whether the target image includes at least two parallel load-bearing beams (2); if two parallel load-bearing beams (2) are identified, the three-dimensional position and posture of each load-bearing beam (2) is identified based on image processing; based on the three-dimensional position and posture data of the two load-bearing beams (2), the position and posture of the load-bearing plane formed by the two load-bearing beams (2) is calculated; and based on the position and posture of the load-bearing plane, the AGV is controlled to accurately place the transformer on the load-bearing plane, and automatically retreat along a preset route after completion.

10. An electronic device, characterized in that: include: Memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the AGV-based intelligent transformer transportation method according to any one of claims 1 to 8 is implemented.

Images