A device and method for adjusting parallelism based on a two-dimensional code

Abstract

The application discloses a device for adjusting operation parallelism based on a two-dimensional code, and relates to the technical field of robots, comprising an AGV module, a mechanical arm module, a terminal module, a visual information processing module, a communication module and a general control module. The application also discloses a method for adjusting operation parallelism based on a two-dimensional code, comprising the following steps: S100, constructing a map; S200, placing a two-dimensional code; S300, coarsely adjusting an observation posture; S400, finely adjusting the observation posture; and S500, adjusting operation parallelism. The application has the characteristics of high stability, strong self-adaptability, wide application range, strong generalization and remarkable effect, and can more efficiently meet the needs of complex scenes.

Description

Technical Field

[0001] This invention relates to the field of robotics, and more particularly to a device and method for adjusting the parallelism of operations based on QR codes. Background Technology

[0002] The next generation of mobile robot systems should possess the ability to interact with complex environments, with perception and motion control being two of the most crucial capabilities. Perception refers to the mobile robot system's ability to acquire information about its external environment through various sensor devices, including cameras and LiDAR. Motion control refers to processing information from external feedback using methods such as deep networks, model building, and real-time mapping. This information serves as the basis for designing actions, enabling the mobile robot to perform movements such as movement, lifting, recognition, and grasping while ensuring real-time interaction with the complex environment. In practical tasks, these actions often need to be designed and combined based on real-world environmental information to achieve given requirements, such as the most basic and core task of a mobile robot system—moving and manipulating target objects on demand. The objective of this task is to move the robot to a designated location as required and interact with the environment, then control the robotic arm and end effector to switch various switches to specified states. The general workflow for a mobile robot can be roughly divided into: controlling the AGV (Automated Guided Vehicle) to identify the environment and perform SLAM (Simultaneous Localization and Mapping) mapping; controlling the AGV to move to the approximate target area; using motion sensing devices to identify target object information, including position, type, and posture; based on the existing information, planning the robotic arm's motion path and designing operational actions, controlling the robotic arm to move to the target object and complete the relevant actions; and operating the next switch. Therefore, accurately controlling the robot system and providing a good information acquisition foundation are necessary conditions for achieving high-precision operation of target objects. Currently, in actual operating environments, the following problems still exist in the high-precision operation of mobile robot systems:

[0003] 1. Although current cutting-edge or open-source SLAM algorithms and coherent navigation algorithms are relatively mature and highly accurate, the errors in both position and angle are amplified when these algorithms are applied to AGVs due to factors such as hardware conditions, system integration, and communication methods. Currently, the actual accuracy deviation of fixed-point navigation for indoor AGVs in China is approximately 0.1-0.25m for position and 0.1-0.5rads for angle. Furthermore, the secondary calibration after each AGV restart also causes a certain deviation in map accuracy, resulting in new errors for the same data.

[0004] 2. From a sensor perspective, although various types of sensors have made some progress, for scenarios where robots interact with the outside world, cameras still provide visual information, and depth recognition networks are used for target identification, which can meet diverse and varied identification needs. However, in practical applications, the recognition effect of existing recognition networks is affected by factors such as the training environment, the actual operating environment, lighting conditions, and the recognition angle. Therefore, when the AGV positioning is inaccurate, environmental conditions such as the recognition angle will change significantly, greatly reducing the recognition effect, especially when there are many and varied types of switches.

[0005] 3. In real-world scenarios, high-precision robot operation is essential for ensuring safety and stability. Taking substation inspection as an example, the electrical equipment operated during the inspection process has a high safety factor, thus requiring high precision and safe operation (such as preventing interference). However, due to the actual navigation accuracy of AGVs, deviations in position and angle can severely affect operational precision and lead to interference with switch operations. Therefore, additional mechanisms or software designs are needed to compensate for this deficiency.

[0006] Therefore, those skilled in the art are dedicated to developing a device and method for adjusting the parallelism of operations based on QR codes. Summary of the Invention

[0007] In view of the above-mentioned deficiencies of the prior art, the technical problem to be solved by the present invention is the instability of the visual recognition environment and the robotic arm operation environment caused by inaccurate AGV navigation.

[0008] The inventors analyzed the impact of inaccurate navigation in existing AGV systems on recognition and operation, which significantly reduces the robotic arm's precision in manipulating target objects. The root causes of navigation inaccuracy lie in factors such as hardware infrastructure, system layout, sensor sensitivity, and transmission latency. Since the algorithms themselves are already relatively accurate and mature, improvements in hardware and systems are not feasible given the same cost constraints. Regarding the issue of unstable recognition angles, environmental changes and angle variations need to be considered, requiring the selection of appropriate deep networks and training on various labeled images. This necessitates substantial time and personnel costs to achieve ideal results, and the effectiveness is further affected by changes in the environment. Finally, the issue of operational interference requires the introduction of additional sensing information devices or complex robotic arm trajectory planning, which also demands significant time and personnel costs.

[0009] The inventors used AGV and SLAM-related information as prior information, along with a QR code placed in a suitable position. They combined a camera and a recognition algorithm to read the three-dimensional spatial relationship between the robotic arm's end effector and the QR code in real time. The two were used together to design an attitude calculation algorithm. The obtained information was used to adaptively adjust the robotic arm's pose in real time, so that the plane where the camera in the end effector module is located is parallel to the plane where the target object is located. This provides a good operating environment and a target recognition environment for the neural network for subsequent visual recognition and robotic arm operation.

[0010] The inventors define the target object as the target of the robotic arm operation, including but not limited to switches, knobs, buttons, and press plates. Generally, the operation of the robotic arm involves multiple target objects.

[0011] In one embodiment of the present invention, an apparatus for adjusting the parallelism of operations based on a QR code is provided, comprising:

[0012] The AGV module performs environmental perception and safe path planning;

[0013] The robotic arm module can be accurately controlled at any angle and position within a given radius to achieve high-precision combined operations.

[0014] The end effector module integrates a gripper, camera, and supplementary lighting equipment. It is connected to the robotic arm module via a linkage structure and interacts directly with external objects.

[0015] The visual information processing module acquires real-time images and depth information, recognizes QR codes, and runs a deep network.

[0016] The communication module uses wireless communication for information exchange between modules.

[0017] The central control module receives motion information sent by the AGV module, robotic arm module, end effector module, and vision information processing module through the communication module, and issues control commands to the AGV module, robotic arm module, end effector module, and vision information processing module through the communication module to perform integrated control and scheduling.

[0018] The AGV module serves as the chassis. The vision information processing module is connected to the AGV module via cables. The robotic arm module is connected to the AGV module via a linkage structure. The end effector module is connected to the robotic arm module via a linkage structure. The vision information processing module is connected to the camera in the end effector module via cables. The AGV module, robotic arm module, end effector module, and vision information processing module communicate wirelessly with the central control module through a communication module, transmitting their respective motion information to the central control module and receiving control commands from the central control module.

[0019] Optionally, in the QR code-based device for adjusting the parallelism of operations in the above embodiments, the AGV module integrates SLAM, obstacle avoidance, automatic recharging, path planning and navigation functions.

[0020] Optionally, in the device for adjusting the parallelism of the operation based on QR code in any of the above embodiments, the AGV module provides an external operation interface, including external operation interfaces for SLAM, navigation, and correction parameters.

[0021] Optionally, in the device for adjusting the parallelism of operations based on QR codes in any of the above embodiments, the AGV module provides a programming interface so that external devices can control the AGV to complete SLAM, navigation, and path planning functions through a programming language.

[0022] Furthermore, in the device for adjusting the parallelism of the operation based on QR codes in the above embodiments, the programming interface includes an API interface, a dynamic link library, a static link library, and a code usage reference.

[0023] Optionally, in the device for adjusting the parallelism of the operation based on QR code in any of the above embodiments, the AGV module is also provided with a power port, a network port, and a USB port so as to connect other devices and provide power supply or connection to the outside world.

[0024] Optionally, in the device for adjusting the parallelism of the operation based on QR codes in any of the above embodiments, the AGV module uses Chuangze's AGV.

[0025] Optionally, in the device for adjusting the parallelism of the operation based on QR code in any of the above embodiments, the robotic arm module integrates modeling, forward and inverse kinematics, motion trajectory planning, and motion dead zone functions.

[0026] Optionally, in the device for adjusting the parallelism of the operation based on QR code in any of the above embodiments, the robotic arm module provides a terminal program to control the robotic arm, realize parameter modification and robotic arm model simulation, the parameter modification includes robotic arm parameter modification, communication parameter modification, and safety protection parameter modification, and the robotic arm model simulation includes robotic arm model display and movement control.

[0027] Furthermore, in the device for adjusting the parallelism of the operation based on QR codes in the above embodiments, the terminal program includes an APP or PC software.

[0028] Optionally, in the QR code-based device for adjusting the parallelism of operations in any of the above embodiments, the robotic arm module provides a programming interface to control the robotic arm to complete the desired movement through a programming language.

[0029] Furthermore, in the device for adjusting the parallelism of the operation based on QR codes in the above embodiments, the programming interface includes an API interface file, a dynamic link library, a static link library, and a code usage reference.

[0030] Optionally, in the device for adjusting the parallelism of the operation based on QR codes in any of the above embodiments, the robotic arm module is a six-axis robotic arm.

[0031] Furthermore, in the device for adjusting the parallelism of the operation based on QR codes in the above embodiments, the robotic arm module is selected as Jaka Zu7.

[0032] Optionally, in the device for adjusting the parallelism of the operation based on the QR code in any of the above embodiments, the visual information processing module runs a program to recognize the QR code and the depth network, and obtains real-time image information and depth information through a wired connection to the camera in the end module.

[0033] Optionally, in the device for adjusting the parallelism of the operation based on QR codes in any of the above embodiments, the visual information processing module uses a Jetson Xavier NX processor.

[0034] Optionally, in the device for adjusting the parallelism of the operation based on the QR code in any of the above embodiments, the camera uses a RealSense L515.

[0035] Optionally, in the device for adjusting the parallelism of the operation based on QR code in any of the above embodiments, the wireless communication method includes ZMQ.

[0036] Optionally, in the device for adjusting the parallelism of the operation based on QR codes in any of the above embodiments, the main control module uses a host computer or a laptop computer.

[0037] Based on any of the above embodiments, in another embodiment of the present invention, a method for adjusting the parallelism of operations based on QR codes is provided, comprising the following steps:

[0038] S100. Construct a map, perform SLAM on the environmental information to obtain a two-dimensional map, and set preset AGV locations and basic information according to the distribution of target objects.

[0039] S200: Place QR codes according to the distribution of the target objects, and set the robotic arm observation posture according to the position of the QR codes.

[0040] S300: Coarse adjustment of observation posture. Based on the preset AGV position and the actual AGV position, coarse adjustment of the robotic arm's observation posture is made to ensure that the QR code is within the field of view.

[0041] S400, fine-tune the observation posture. Use the QR code to fine-tune the observation posture of the robotic arm.

[0042] S500: Adjust the parallelism of the operation, calculate the vector information between the center of the camera and the center of the QR code in the end module. If it is less than or equal to the given threshold, then the plane where the camera is located in the end module is parallel to the plane where the target object is located; otherwise, repeat S400.

[0043] Optionally, in the method for adjusting the parallelism of operations based on QR codes in the above embodiments, step S100 includes:

[0044] S110: Control the movement of the AGV, collect environmental information, and create a two-dimensional map;

[0045] S120. Set preset AGV points. Based on the actual site conditions, control the AGV to move in front of the target object to ensure that the robotic arm can control the target object. Set the preset AGV points according to this standard.

[0046] S130. Measure planar angle information, control AGV movement, and measure the basic information of the target object in the current two-dimensional map, including planar angle information.

[0047] Optionally, in the method for adjusting the parallelism of operations based on QR codes in the above embodiments, step S200 includes:

[0048] S210. Place the QR code in an area where the target objects are densely distributed, according to the distribution of the target objects.

[0049] S220, coarse adjustment error, control the AGV to navigate to each preset AGV point in sequence, and manually compensate and adjust the actual position error and actual angle error of the AGV according to the information of the preset AGV point to keep the actual AGV point and the preset AGV point information the same.

[0050] S230. Set the observation posture. Based on the different preset AGV points and the location of the QR code, set the observation posture of the robotic arm.

[0051] Furthermore, in the method for adjusting the parallelism of the operation based on the QR code in the above embodiments, the QR code has a size of 10cm*10cm.

[0052] Optionally, in the method for adjusting the parallelism of operations based on QR codes in the above embodiments, step S300 includes:

[0053] S310. Obtain information on the actual AGV location reached. Based on the target object, control the corresponding preset AGV location. After navigation is completed, obtain information on the actual AGV location reached.

[0054] S320: Adjust the observation posture, combine the preset AGV position and basic information with the actual AGV position information, use two-dimensional geometry to calculate the position deviation information and angle deviation, and adjust the robotic arm observation posture.

[0055] Optionally, in the method for adjusting the parallelism of operations based on QR codes in any of the above embodiments, step S400 includes:

[0056] S410, Observe the QR code. Move the robotic arm to the adjusted observation posture and observe the QR code.

[0057] S420. Solve the angle deviation. Based on the information of the four corner points of the QR code and the visual depth information of the corresponding pixels, calculate the angle deviation in the three angular directions of the Euler angles in turn.

[0058] S430, Compensation deviation, based on the coordinate axis of the camera in the end module, compensates for the angle deviation, and then compensates it to the robot arm pose through matrix operation, thereby adjusting the robot arm's observation posture.

[0059] Optionally, in the method for adjusting the parallelism of operations based on QR codes in any of the above embodiments, step S500 includes:

[0060] S510. Adjust the robotic arm to recognize the QR code in the coarsely adjusted observation posture. Based on the recognized Cartesian space coordinates, control the movement of the robotic arm to place the QR code at the visual center.

[0061] S520. Calculate the deviation, recognize the QR code again, calculate the vector information between the center of the camera in the end module and the center of the QR code. If it is less than or equal to the given threshold, the plane where the camera is located in the end module and the plane where the target object is located are considered to be parallel; execute step S400 again.

[0062] This invention, based on existing information and simple QR code labels, compensates for the inaccuracy and instability of navigation. By utilizing geometric and three-dimensional spatial structural relationships and mathematical analysis, it compensates for the errors brought by the AGV into the pose of the robotic arm, achieving the goal of not affecting recognition and operation. It features high stability, strong adaptability, wide applicability, strong generalization, and significant effects, and can more efficiently meet the needs of complex scenarios.

[0063] The following will further explain the concept, specific structure, and technical effects of the present invention in conjunction with the accompanying drawings, so as to fully understand the purpose, features, and effects of the present invention. Attached Figure Description

[0064] Figure 1 This is a schematic diagram illustrating the structure of a device for adjusting the parallelism of operations based on a QR code, according to an exemplary embodiment.

[0065] Figure 2 This is a flowchart illustrating a method for adjusting operational parallelism based on a QR code, according to an exemplary embodiment. Detailed Implementation

[0066] The following description, with reference to the accompanying drawings, illustrates several preferred embodiments of the present invention to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.

[0067] In the accompanying drawings, components with the same structure are indicated by the same numerical designation, and components with similar structures or functions are indicated by similar numerical designations. The dimensions and thicknesses of each component shown in the drawings are arbitrary, and the present invention does not limit the dimensions and thicknesses of each component. To make the illustrations clearer, the thickness of components is schematically exaggerated in some places in the drawings.

[0068] The inventors designed a device based on QR codes to adjust the parallelism of operations, such as... Figure 1 As shown, it includes:

[0069] The AGV module performs environmental perception and safe path planning; it integrates SLAM, obstacle avoidance, automatic recharging, path planning, and navigation functions; it provides external operation interfaces, including interfaces for SLAM, navigation, and parameter correction; it provides a programming interface so that external devices can control the AGV to complete SLAM, navigation, and path planning functions through programming languages. The programming interface includes API interfaces, dynamic link libraries, static link libraries, and code usage references; the AGV module also has a power port, network port, and USB port for mounting other devices and providing power supply or connectivity; preferably, the AGV module uses AGVs from Chuangze.

[0070] The robotic arm module enables precise control at any angle and position within a given radius, achieving high-precision combined operations. It integrates modeling, forward and inverse kinematics, motion trajectory planning, and dead-zone detection. A terminal program is provided to control the robotic arm, allowing for parameter modification and model simulation. Parameter modification includes changes to robotic arm parameters, communication parameters, and safety protection parameters. Model simulation includes display and movement control of the robotic arm model. The terminal program includes an app or PC software. A programming interface is provided, including API files, dynamic link libraries, static link libraries, and code usage references, allowing control of the robotic arm through a programming language to achieve desired movements. The robotic arm module is a six-axis robotic arm, preferably based on the Jaka Zu7.

[0071] The end effector module integrates a gripper, camera, and supplementary lighting equipment. It is connected to the robotic arm module via a linkage structure and interacts directly with external objects. The preferred camera is the RealSense L515.

[0072] The visual information processing module acquires real-time images and depth information, recognizes QR codes, and runs a deep network; the program that runs the QR code recognition and deep network is connected to the camera in the end module via a wired connection to acquire real-time image and depth information; the visual information processing module preferably uses a Jetson Xavier NX processor;

[0073] The communication module uses wireless communication methods for information exchange between modules, including ZMQ.

[0074] The central control module receives motion information from the AGV module, robotic arm module, end effector module, and vision information processing module via the communication module, and issues control commands to the AGV module, robotic arm module, end effector module, and vision information processing module via the communication module for comprehensive control and scheduling; the central control module uses a host computer or laptop computer.

[0075] The AGV module serves as the chassis. The vision information processing module is connected to the AGV module via cables. The robotic arm module is connected to the AGV module via a linkage structure. The end effector module is connected to the robotic arm module via a linkage structure. The vision information processing module is connected to the camera in the end effector module via cables. The AGV module, robotic arm module, end effector module, and vision information processing module communicate wirelessly with the central control module through a communication module, transmitting their respective motion information to the central control module and receiving control commands from the central control module.

[0076] Based on the above embodiments, the inventors provide a method for adjusting the parallelism of operations based on QR codes, such as... Figure 2 As shown, it includes the following steps:

[0077] S100. Construct a map, perform SLAM on the environmental information to obtain a two-dimensional map, and set preset AGV locations and basic information based on the distribution of target objects; specifically including:

[0078] S110: Control the movement of the AGV, collect environmental information, and create a two-dimensional map;

[0079] S120. Set preset AGV points. Based on the actual site conditions, control the AGV to move in front of the target object to ensure that the robotic arm can control the target object. Set the preset AGV points according to this standard.

[0080] S130. Measure planar angle information, control AGV movement, and measure the basic information of the target object in the current two-dimensional map, including planar angle information.

[0081] S200: Place QR codes. Based on the distribution of the target objects, place QR codes and set the robotic arm's observation posture according to the QR code positions; specifically including:

[0082] S210. Place the QR code. Based on the distribution of the target objects, place the QR code in the area where the target objects are densely distributed. The preferred size of the QR code is 10cm*10cm.

[0083] S220, coarse adjustment error, control the AGV to navigate to each preset AGV point in sequence, and manually compensate and adjust the actual position error and actual angle error of the AGV according to the information of the preset AGV point to keep the actual AGV point and the preset AGV point information the same.

[0084] S230, Set the observation posture, based on the different AGV locations and the location of the QR code.

[0085] Set the robotic arm's observation posture.

[0086] S300, coarse adjustment of observation posture: Based on the preset AGV position and the actual AGV position, coarsely adjust the robotic arm's observation posture to ensure the QR code is within the field of view; specifically including:

[0087] S310. Obtain information on the actual AGV location reached. Based on the target object, control the corresponding preset AGV location. After navigation is completed, obtain information on the actual AGV location reached.

[0088] S320: Adjust the observation posture, combine the preset AGV position and basic information with the actual AGV position information, use two-dimensional geometry to calculate the position deviation information and angle deviation, and adjust the robotic arm observation posture.

[0089] S400, fine-tuning the observation attitude: Based on the QR code, fine-tune the robotic arm's observation attitude, specifically including:

[0090] S410, Observe the QR code. Move the robotic arm to the adjusted observation posture and observe the QR code.

[0091] S420. Solve the angle deviation. Based on the information of the four corner points of the QR code and the visual depth information of the corresponding pixels, calculate the angle deviation in the three angular directions of the Euler angles in turn.

[0092] S430, Compensation deviation, based on the coordinate axis of the camera in the end module, compensates for the angle deviation, and then compensates it to the robot arm pose through matrix operation, thereby adjusting the robot arm's observation posture.

[0093] S500: Adjust the parallelism of the operation, calculate the vector information between the center of the camera and the center of the QR code in the end module. If it is less than or equal to a given threshold, then the plane where the camera is located in the end module is parallel to the plane where the target object is located; otherwise, repeat S400. Specifically, this includes:

[0094] S510. Adjust the robotic arm to recognize the QR code in the coarsely adjusted observation posture. Based on the recognized Cartesian space coordinates, control the movement of the robotic arm to place the QR code at the visual center.

[0095] S520. Calculate the deviation, recognize the QR code again, calculate the vector information between the center of the camera in the end module and the center of the QR code. If it is less than or equal to the given threshold, the plane where the camera is located in the end module and the plane where the target object is located are considered to be parallel; execute step S400 again.

[0096] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for adjusting the parallelism of operations based on QR codes, using a device for adjusting the parallelism of operations based on QR codes, comprising: The AGV module performs environmental perception and safe path planning; The robotic arm module can be accurately controlled at any angle and position within a given radius to achieve high-precision combined operations. The end effector module integrates a gripper, camera, and supplementary lighting equipment, and is connected to the robotic arm module through a linkage structure, allowing it to interact directly with external objects; The visual information processing module acquires real-time images and depth information, recognizes QR codes, and runs a deep network. The communication module uses wireless communication for information exchange between modules. The central control module receives motion information sent by the AGV module, the robotic arm module, the end effector module, and the vision information processing module through the communication module, and issues control commands to the AGV module, the robotic arm module, the end effector module, and the vision information processing module through the communication module to perform integrated control scheduling. The AGV module is a chassis. The vision information processing module is connected to the AGV module via a cable. The robotic arm module is connected to the AGV module via a linkage structure. The vision information processing module is connected to the camera in the end effector module via a cable. The AGV module, the robotic arm module, the end effector module, and the vision information processing module communicate wirelessly with the central control module through the communication module, transmitting their respective motion information to the central control module and receiving control commands from the central control module. Its features include the following steps: S100. Map Building: Perform SLAM on the environmental information to obtain a two-dimensional map, and set preset AGV locations and basic information based on the distribution of target objects; specifically including: S110. Control the AGV to move, collect environmental information, and establish the two-dimensional map; S120. Set preset AGV positions: Based on the actual site conditions, control the AGV to move in front of the target object to ensure that the robotic arm can manipulate the target object. Set the preset AGV positions according to this standard. S130, Measure planar angle information: Control the AGV to move and measure the basic information of the target object in the current two-dimensional map, including planar angle information; S200, Placing QR Codes: Based on the distribution of the target objects, QR codes are placed, and the robotic arm's observation posture is set according to the QR code positions; specifically including: S210. Placing the QR code: Based on the distribution of the target objects, place the QR code in the area where the target objects are densely packed; S220, Coarse adjustment error: Control the AGV to navigate to each preset AGV point in sequence, and manually compensate and adjust the actual position error and actual angle error of the AGV according to the information of the preset AGV point, so as to keep the actual AGV point and the preset AGV point information the same. S230. Set observation posture: Set the observation posture of the robotic arm according to the different preset AGV points and the location of the QR code; S300, Coarse adjustment of observation posture: Based on the preset AGV position and the actual AGV position, coarsely adjust the observation posture of the robotic arm to ensure that the QR code is within the field of view; S400. Fine-tune the observation posture: Fine-tune the observation posture of the robotic arm according to the QR code. S500, Adjusting Operation Parallelism: Calculate the vector information between the center of the camera in the end module and the center of the QR code. If it is less than or equal to a given threshold, then the plane where the camera in the end module is located is parallel to the plane where the target object is located; otherwise, repeat S400.

2. The method for adjusting the parallelism of operations based on QR codes as described in claim 1, characterized in that, Step S300 includes: S310. Obtain information on the actual AGV location reached: Based on the target object, control the corresponding preset AGV location, and after navigation is completed, obtain information on the actual AGV location reached. S320. Adjust the observation posture: Combining the preset AGV position with the basic information and the actual AGV position information, use two-dimensional geometry to calculate the position deviation information and angle deviation, and adjust the observation posture of the robotic arm.

3. The method for adjusting the parallelism of operations based on QR codes as described in claim 2, characterized in that, Step S400 includes: S410, Observe the QR code: Move the robotic arm to the adjusted observation posture and observe the QR code; S420. Calculate the angle deviation: Based on the information of the four corner points of the QR code and the visual depth information of the corresponding pixels, calculate the angle deviation in the three angular directions of the Euler angles in sequence. S430, Compensation for Deviation: Based on the coordinate axis of the camera in the end module, the angle deviation is compensated, and then the compensation is applied to the pose of the robotic arm through matrix operation to adjust the observation posture of the robotic arm.

4. The method for adjusting the parallelism of operations based on QR codes as described in claim 3, characterized in that, Step S500 includes: S510. Adjust the robotic arm to recognize the QR code in the coarsely adjusted observation posture. Based on the recognized Cartesian space coordinates, control the movement of the robotic arm so that the QR code is located at the visual center. S520. Calculate the deviation, re-identify the QR code, calculate the vector information between the center of the camera in the end module and the center of the QR code, if it is less than or equal to the given threshold, then the plane where the camera in the end module is located and the plane where the target object is located are considered to be parallel; execute step S400 again.

Images