Hand-eye automatic calibration robot system, calibration method and related device

Abstract

The invention discloses a robot system with automatic hand-eye calibration, a calibration method and a related device, and belongs to the technical field of intelligent robots with bodies, the robot system comprises a multi-axis robot, and the multi-axis robot moves and works in a working area; the hand-eye calibration tool has the same degree of freedom of motion as that of the multi-axis robot; the visual sensor is mounted on an executing mechanism of the multi-axis robot and is used for acquiring point cloud data information of the hand-eye calibration tool in a visual field of the visual sensor; the information processing computer is used for receiving the point cloud data information acquired by the visual sensor, identifying the pose coordinate information of the hand-eye calibration tool and carrying out feature calculation to obtain feature coordinate information; and planning a motion path of the multi-axis robot and a measurement position of the visual sensor according to the characteristic coordinate information, and controlling the multi-axis robot to execute a pre-planned path sequence. According to the invention, the motion path and the shooting position required by calibration can be automatically planned through the rough position of the calibration tool, and the calibration work can be simply, accurately and efficiently completed.

Description

Technical Field

[0001] This invention belongs to the field of embodied intelligent robot technology, specifically relating to a robot system, calibration method and related device for automatic hand-eye calibration. Background Technology

[0002] In existing robotic systems, the relative positional relationship between the vision sensor and the robot system is typically established through manual teaching. This involves manually moving a tool point (TCP) mounted on the robot to the target point to obtain the coordinates of that point in both the robot's coordinate system and the sensor's coordinate system. This process requires manual judgment of the robot's positioning, which is not only time-consuming but also has inconsistent accuracy, relying heavily on the operator's experience. Furthermore, when a collision occurs or the system's installation state changes, recalibration is necessary, a highly cumbersome process that significantly increases labor costs.

[0003] Existing technologies, such as patent application CN110497386B, disclose an "automatic hand-eye relationship calibration method for collaborative robots," which includes: placing a calibration board; guiding the collaborative robot to a teaching point and recording the robot's coordinates; guiding the collaborative robot to an initial photo point, recording the robot's coordinates, and configuring a matching template; controlling the robot to continuously change the photo point to obtain robot coordinates and marker pixel coordinate point pairs; and calculating calibration parameters. Another example is patent application CN113664835A, which discloses an "automatic hand-eye calibration method for robots." This method automatically plans a path by manually setting the initial calibration position value and the search ball space step size. The robotic arm uses an automatic path planning method, planning the path according to preset search constraints, and calculating multiple sets of photo points. For example, patent application CN118906064A discloses "An Automatic Hand-Eye Calibration Method for Delta Robots." The calibration board is located on the Delta robot's moving platform and has a color dot matrix. Multiple point pairs are formed using the spatial coordinates and image coordinates of the obtained color dot matrix centers, and the mapping matrix is ​​estimated, simplifying the hand-eye calibration process and improving work efficiency. Another example is patent application CN115781698A, which discloses "A Layered Automatic Generation Method for Robot Motion Pose in Hand-Eye Calibration." Based on the identification and extraction of the calibration board's pixel dimensions (length and width), the current physical pixel conversion coefficient is obtained, along with the maximum distance the robot can move. Based on preset layers, layer height, layer length and width, the number of points in the length direction, the number of points in the width direction, and the maximum distance the robot can move, a set of Cartesian points representing the robot's relative motion is obtained. The number of movement points is manually constrained during the process. The existing methods described above employ pre-set template parameters, step sizes, or manual guidance to enable the robot to acquire the feature information of the calibration fixture. The transition paths during the process require manual teaching. Compared to traditional calibration methods, these methods only simplify the calibration process and are not fully automated. Furthermore, the calibration fixture must conform to the system's recognition parameters and possess a specific color or shape, rather than being arbitrary; the process lacks intelligence. Summary of the Invention

[0004] The purpose of this invention is to address the problems in the prior art by providing a robot system, calibration method, and related devices for automatic hand-eye calibration. This system can automatically plan the motion path and shooting position required for calibration based on the rough position of the calibration fixture. Through automatic planning, automatic execution, and automatic calculation, the calibration relationship between the robot and the vision sensor is obtained, thereby enabling the calibration work to be completed simply, accurately, and efficiently.

[0005] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, a robotic system for automatic hand-eye alignment is provided, comprising: Multi-axis robots perform tasks within a work area; The hand-eye calibration fixture has the same degrees of freedom of movement as a multi-axis robot; A vision sensor is installed on the actuator of a multi-axis robot to acquire point cloud data information of the hand-eye calibration fixture under the field of view of the vision sensor; The information processing computer receives point cloud data from the vision sensor, identifies the pose coordinates of the hand-eye calibration fixture, performs feature calculations to obtain feature coordinates, and plans the motion path of the multi-axis robot and the measurement position of the vision sensor based on the feature coordinates, controlling the multi-axis robot to execute the pre-planned path sequence.

[0006] As a preferred embodiment, the multi-axis robot has at least three degrees of freedom of motion, and its execution accuracy has been calibrated.

[0007] As a preferred embodiment, the hand-eye calibration fixture has known approximate pose coordinate information. The accuracy range of the approximate pose coordinates deviates from the actual pose coordinates, and the deviation is less than the field of view of the vision sensor.

[0008] Secondly, a calibration method for a robot system based on the aforementioned hand-eye automatic calibration is provided, comprising the following steps: Place the hand-eye calibration fixture within the working area of ​​the multi-axis robot; Obtain approximate pose coordinate information of the hand-eye calibration fixture; the accuracy range of the approximate pose coordinates deviates from the actual pose coordinates, and the deviation is less than the field of view of the vision sensor. The information processing computer performs path planning based on the approximate pose coordinates of the hand-eye calibration fixture; The multi-axis robot moves to the target acquisition and measurement position of the vision sensor according to the path planned by the information processing computer, and acquires the point cloud data information of the hand-eye calibration tool at the target acquisition and measurement position; The information processing computer acquires point cloud data information collected by the vision sensor, and obtains the feature coordinates of the hand-eye calibration fixture in the vision sensor coordinate system and the acquisition pose in the multi-axis robot coordinate system. If the number of collected feature coordinates and the number of multi-axis robot coordinates meet the calculation requirements, then calibration calculation will be performed. The information processing computer calculates the calibration relationship between the multi-axis robot and the vision sensor by solving the set of feature coordinates obtained from the calibration calculation and the set of coordinates of the multi-axis robot. The calibration process ends.

[0009] As a preferred approach, the approximate pose coordinates of the hand-eye calibration fixture are obtained by converting them into a multi-axis robot coordinate system using a third-party sensor for positioning; or, they are obtained by parsing the feature coordinates of the hand-eye calibration fixture.

[0010] As a preferred embodiment, in the step where the multi-axis robot moves to the target acquisition and measurement position of the vision sensor according to the path planned by the information processing computer, the transition path executed by the multi-axis robot to reach the target acquisition and measurement position is calibrated.

[0011] As a preferred embodiment, the steps of the information processing computer acquiring point cloud data information collected by the vision sensor and obtaining the feature coordinates of the hand-eye calibration fixture in the vision sensor coordinate system and the acquired pose in the multi-axis robot coordinate system include: A filtering method is used to remove interference noise from the point cloud data information collected by the visual sensor; Using structured representation and semantic segmentation methods for point cloud information, structural relationships and representation networks containing only the features to be identified by the hand-eye calibration tool are extracted from point cloud data information after removing interference and noise. Based on the structural relationship and representation network containing only the features to be identified by the hand-eye calibration fixture, model parameters of the measured hand-eye calibration fixture are generated. Based on the model parameters of the measured hand-eye calibration fixture, the position and attitude information of the hand-eye calibration fixture in the visual sensor coordinate system are obtained. The pose information of the hand-eye calibration fixture in the vision sensor coordinate system and the pose information of the current multi-axis robot coordinate system are recorded for calibration calculation.

[0012] As a preferred embodiment, the calibration calculation steps include establishing the relationship between corresponding points in three-dimensional space in the visual sensor coordinate system and the multi-axis robot coordinate system, expressed as follows:

[0013] In the formula, For acquiring the pose of a multi-axis robot in a coordinate system, To determine the posture of the hand-eye calibration tooling under the corresponding acquired posture of the multi-axis robot. Let be the pose transformation matrix between the multi-axis robot and the vision sensor coordinate system to be solved; To acquire position data for a multi-axis robot, To determine the position of the hand-eye calibration fixture at the corresponding acquisition position of the multi-axis robot. Here is the position transformation matrix between the multi-axis robot and the vision sensor coordinate system, which needs to be solved. Using the pose information of the hand-eye calibration fixture in the vision sensor coordinate system and the acquired pose information in the current multi-axis robot coordinate system, the pose transformation matrix between the vision sensor coordinate system and the multi-axis robot coordinate system is obtained by using the least squares method.

[0014] As a preferred embodiment, when the information processing computer calculates the calibration relationship between the multi-axis robot and the vision sensor by solving the solution set of the number of feature coordinates obtained through calibration calculation and the number of coordinates of the multi-axis robot, it obtains the rotation relationship of the vision sensor coordinate system relative to the multi-axis robot coordinate system. Peaceful Relationship This allows us to obtain the data transformation relationship between the two coordinate systems.

[0015] Thirdly, a computer-readable storage medium is provided, wherein at least one instruction is stored therein, the at least one instruction being executed by a processor in an electronic device to implement the calibration method described above.

[0016] Compared with the prior art, the present invention has at least the following beneficial effects: This invention uses hand-eye calibration to determine the approximate pose coordinates of the tooling, automatically plans the transition path, automatically identifies the feature positions of the calibration tooling, and automatically calculates the calibration results. This replaces the original manual point-selection calibration process, improving the automation level of the calibration process and reducing manual time consumption. It does not require highly skilled operators; the system automatically obtains the calibration results of the visual sensor coordinate system relative to the multi-axis robot coordinate system. The tooling feature information is identified and extracted using point cloud data, and the feature pose coordinates in the sensor point cloud coordinate system are transformed to the multi-axis robot coordinate system. The entire process is executed automatically without human intervention and does not require complex operator skills. The calibration method of this invention is simple and efficient, the calibration process is automatically executed, no complex manual operations are required, and the accuracy is stable.

[0017] Furthermore, the hand-eye calibration fixture of this invention has known approximate pose coordinate information. The accuracy range of the approximate pose coordinates deviates from the actual pose coordinates, and this deviation is smaller than the field of view of the vision sensor. The approximate pose coordinate information of the hand-eye calibration fixture of this invention is obtained by localization through recognition and transformation to a multi-axis robot coordinate system by a third-party sensor; or by parsing the feature coordinate information of the hand-eye calibration fixture. This is compatible with system structures that include localization transformation using a third-party vision sensor, as well as system structures that do not include a third-party vision sensor and rely on a system model for localization.

[0018] Furthermore, by ensuring that all motion paths executed during the automatic calibration process are automatically planned non-interference paths, the present invention reduces the degree of human intervention and time consumption in the calibration process by eliminating the need for manual setting of calibration acquisition points and process path points.

[0019] Furthermore, this invention employs a filtering method to remove interference noise from the point cloud data information collected by the visual sensor, and then segments and extracts it using a semantic segmentation method. The feature poses of corresponding points in the point cloud are identified through a feature recognition algorithm, and the feature poses in the sensor point cloud coordinate system are transformed into the multi-axis robot coordinate system. The feature coordinate system information in the automatic calibration process is automatically calculated by the algorithm, which improves the accuracy and stability of the coordinate points.

[0020] Furthermore, this invention improves the accuracy of the feature transformation matrix by identifying the hand-eye calibration fixture feature poses under different multi-axis robot acquisition poses and using the least squares method to calculate the pose transformation matrix between the vision sensor coordinate system and the multi-axis robot coordinate system for the feature coordinate sequence pairs of corresponding points. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the following drawings are only some of the embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these drawings without creative effort.

[0022] Figure 1 A schematic diagram of the structure of the robot system for automatic hand-eye calibration according to an embodiment of the present invention; Figure 2 Flowchart of a robot system calibration method based on automatic hand-eye calibration according to an embodiment of the present invention; In the attached diagram: 101 - Information processing computer; 102 - Multi-axis robot; 103 - Vision sensor; 104 - Hand-eye calibration fixture. Detailed Implementation

[0023] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0024] Please see Figure 1 This invention proposes a robotic system for automatic hand-eye calibration, aiming to overcome the shortcomings of current calibration methods that require manual aiming and point-by-point measurement, which are time-consuming, inefficient, and demand high operator skill levels. The system includes: The multi-axis robot 102 performs motion work within the work area; The hand-eye calibration fixture 104 has the same degrees of freedom of motion as the multi-axis robot 102; A vision sensor 103 is installed on the actuator of a multi-axis robot 102 to acquire point cloud data information of the hand-eye calibration fixture 104 under the field of view of the vision sensor 103; The information processing computer 101 receives point cloud data information acquired by the vision sensor 103, identifies the pose coordinate information of the hand-eye calibration fixture 104, performs feature calculation to obtain feature coordinate information, and plans the motion path of the multi-axis robot 102 and the measurement position of the vision sensor 103 based on the feature coordinate information, and controls the multi-axis robot 102 to execute the pre-planned path sequence.

[0025] In one possible implementation, the multi-axis robot 102 of the present invention has at least three degrees of freedom of motion, and its execution accuracy has been calibrated. In this embodiment, a six-axis robot is used.

[0026] In one possible implementation, the hand-eye calibration fixture 104 has known approximate pose coordinate information. The accuracy range of the approximate pose coordinates deviates from the actual pose coordinates, and this deviation is smaller than the field of view of the vision sensor 103.

[0027] Please see Figure 2 Another embodiment of the present invention also proposes a calibration method for a robot system based on the aforementioned hand-eye automatic calibration, comprising the following steps: S1. Place the hand-eye calibration fixture 104 in the working area of ​​the multi-axis robot 102; S2. Obtain approximate pose coordinate information of the hand-eye calibration fixture 104; the accuracy range of the approximate pose coordinates deviates from the actual pose coordinates, and the deviation is less than the field of view of the vision sensor 103. S3. The information processing computer 101 performs path planning based on the approximate pose coordinates of the hand-eye calibration fixture 104. S4. The multi-axis robot 102 moves to the target acquisition and measurement position of the vision sensor 103 according to the path planned by the information processing computer 101, and acquires the point cloud data information of the hand-eye calibration fixture 104 at the target acquisition and measurement position. S5. The information processing computer 101 acquires the point cloud data information collected by the vision sensor 103, and obtains the feature coordinates of the hand-eye calibration fixture 104 in the coordinate system of the vision sensor 103 and the acquisition pose in the coordinate system of the multi-axis robot 102. S6. If the number of collected feature coordinates and the number of coordinates of the multi-axis robot 102 meet the calculation requirements, then perform calibration calculation. S7. The information processing computer 101 calculates the calibration relationship between the multi-axis robot 102 and the vision sensor 103 by solving the set of feature coordinates obtained by calibration calculation and the set of coordinates of the multi-axis robot 102. The calibration process ends.

[0028] In one possible implementation, step S1 involves manually placing the hand-eye calibration fixture 104 within the working area of ​​the multi-axis robot 102; step S2 involves obtaining the approximate pose coordinate information of the hand-eye calibration fixture 104 by converting it to the coordinate system of the multi-axis robot 102 through a third-party sensor for localization; or, the information can be obtained by parsing the feature coordinate information of the hand-eye calibration fixture 104 on the system model.

[0029] In one possible implementation, in step S4, the transition path executed by the multi-axis robot 102 to reach the target acquisition and measurement position is calibrated. All motion paths are automatically planned, non-interference paths, eliminating the need for manual setting of calibration acquisition points and process path points, thus reducing the degree of manual intervention and time consumption in the calibration process.

[0030] In one possible implementation, step S5, obtaining the feature coordinates of the hand-eye calibration fixture 104 in the coordinate system of the vision sensor 103 and the acquired pose in the coordinate system of the multi-axis robot 102, includes: A filtering method is used to remove interference noise in the point cloud data information collected by the visual sensor 103; Using structured representation and semantic segmentation methods for point cloud information, structural relationships and representation networks containing only the hand-eye calibration fixture 104 features to be identified are extracted from point cloud data information after removing interference and noise. Based on the structural relationship and representation network containing only the features to be identified in the hand-eye calibration fixture 104, model parameters of the measured hand-eye calibration fixture 104 are generated. Based on the model parameters of the measured hand-eye calibration fixture 104, the position and attitude information of the hand-eye calibration fixture 104 in the coordinate system of the vision sensor 103 are obtained. The pose information of the hand-eye calibration fixture 104 in the coordinate system of the vision sensor 103 and the pose information of the current multi-axis robot 102 are recorded for calibration calculation.

[0031] In one possible implementation, the calibration calculation in step S6 includes establishing the relationship between corresponding points in the three-dimensional space in the coordinate system of the vision sensor 103 and the coordinate system of the multi-axis robot 102, as expressed below:

[0032] In the formula, For acquiring the pose of the multi-axis robot in the 102 coordinate system, To determine the posture of the hand-eye calibration fixture 104 under the corresponding acquired posture of the multi-axis robot 102. The pose transformation matrix between the coordinate system of the multi-axis robot 102 and the vision sensor 103 is to be solved. To acquire the position of the multi-axis robot 102 To determine the position of the hand-eye calibration fixture 104 at the corresponding acquisition position of the multi-axis robot 102. The position transformation matrix between the coordinate systems of the multi-axis robot 102 and the vision sensor 103 is to be solved. Using the pose information of the hand-eye calibration fixture 104 in the coordinate system of the vision sensor 103 and the acquired pose information in the coordinate system of the current multi-axis robot 102, the pose transformation matrix between the coordinate system of the vision sensor 103 and the coordinate system of the multi-axis robot 102 is obtained by using the least squares method.

[0033] Furthermore, step S7 obtains the rotational relationship between the coordinate system of the vision sensor 103 and the coordinate system of the multi-axis robot 102. Peaceful Relationship This allows us to obtain the data transformation relationship between the two coordinate systems.

[0034] In one possible implementation, if the number of feature coordinates collected in step S6 does not meet the calculation requirements of the number of coordinates of the multi-axis robot 102, then steps S3 to S5 are repeated, that is, the process of the information processing computer 101 planning the path, the vision sensor 103 collecting point cloud data, and the information processing computer 101 identifying features is repeated.

[0035] This invention discloses an automatic hand-eye calibration method for a robot system, which calibrates the installation relationship between the robot and the vision sensor. This method is automated with low human intervention. It employs structured representation and semantic segmentation of point cloud information, eliminating the need for fixed calibration fixtures and allowing for the identification and representation of any fixture. Based on the approximate pose of the known calibration fixture, it automatically plans the non-interference transition path and posture required for calibration, as well as the precise positioning and shooting posture of the sensor. The robot executes the planning results, the sensor collects point cloud data, and the topological network of the calibration fixture's features is identified through the point cloud data, thereby identifying the fixture's pose and calculating the necessary calibration data. Simultaneously, if the number of collected feature coordinates does not meet the calculation requirements compared to the number of multi-axis robot coordinates, the information processing computer re-plans the path, the vision sensor collects point cloud data, and the information processing computer identifies the features based on the fixture's feature pose information, all without human intervention.

[0036] Another embodiment of the present invention provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the calibration method described above.

[0037] For example, the instructions stored in the memory can be divided into one or more modules / units. These modules / units are stored in a computer-readable storage medium and executed by the processor to complete the calibration method described in this embodiment of the invention. The one or more modules / units can be a series of computer-readable instruction segments capable of performing a specific function, which describe the execution process of the computer program in the server.

[0038] The electronic device may be a smartphone, laptop, PDA, or cloud server, among other computing devices. It may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the electronic device may also include more or fewer components, or combinations of certain components, or different components; for example, it may also include input / output devices, network access devices, buses, etc.

[0039] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0040] The memory can be an internal storage unit of the server, such as the server's hard drive or memory. The memory can also be an external storage device of the server, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card.

[0041] Furthermore, the memory may include both internal storage units of the server and external storage devices. The memory is used to store the computer-readable instructions and other programs and data required by the server. The memory can also be used to temporarily store data that has been output or will be output.

[0042] It should be noted that the information interaction and execution process between the above-mentioned module units are based on the same concept as the method embodiment. For details on their specific functions and technical effects, please refer to the method embodiment section. They will not be repeated here.

[0043] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0044] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0045] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0046] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A robot system for automatic hand-eye calibration, characterized in that, include: A multi-axis robot (102) performs motion work within the work area; The hand-eye calibration fixture (104) has the same degrees of freedom of motion as the multi-axis robot (102); A vision sensor (103) is installed on the actuator of a multi-axis robot (102) to acquire point cloud data information of the hand-eye calibration fixture (104) under the field of view of the vision sensor (103); The information processing computer (101) receives point cloud data information acquired by the vision sensor (103), identifies the pose coordinate information of the hand-eye calibration fixture (104), performs feature calculation to obtain feature coordinate information, and plans the motion path of the multi-axis robot (102) and the measurement position of the vision sensor (103) according to the feature coordinate information, and controls the multi-axis robot (102) to execute the pre-planned path sequence.

2. The robot system for automatic hand-eye calibration according to claim 1, characterized in that, The multi-axis robot (102) has at least three degrees of freedom of motion, and its execution accuracy has been calibrated.

3. The robot system for automatic hand-eye calibration according to claim 1, characterized in that, The hand-eye calibration fixture (104) has known approximate pose coordinate information. The accuracy range of the approximate pose coordinates deviates from the actual pose coordinates. The deviation is less than the field of view of the vision sensor (103).

4. A calibration method for a robot system based on the automatic hand-eye calibration described in any one of claims 1 to 3, characterized in that, Includes the following steps: Place the hand-eye calibration fixture (104) in the working area of ​​the multi-axis robot (102); Obtain approximate pose coordinate information of the hand-eye calibration fixture (104); the accuracy range of the approximate pose coordinates deviates from the actual pose coordinates, and the deviation is less than the field of view of the vision sensor (103). The information processing computer (101) performs path planning based on the approximate pose coordinate information of the hand-eye calibration fixture (104); The multi-axis robot (102) moves to the target acquisition and measurement position of the vision sensor (103) according to the path planned by the information processing computer (101), and acquires the point cloud data information of the hand-eye calibration fixture (104) at the target acquisition and measurement position; The information processing computer (101) acquires the point cloud data information collected by the vision sensor (103) and obtains the feature coordinates of the hand-eye calibration fixture (104) in the coordinate system of the vision sensor (103) and the acquisition pose in the coordinate system of the multi-axis robot (102). If the number of feature coordinates collected meets the calculation requirements of the number of coordinates of the multi-axis robot (102), then calibration calculation is performed. The information processing computer (101) calculates the calibration relationship between the multi-axis robot (102) and the vision sensor (103) by solving the set of feature coordinates obtained by calibration calculation and the coordinates of the multi-axis robot (102), and the calibration process ends.

5. The calibration method according to claim 4, characterized in that, The approximate pose coordinate information of the hand-eye calibration fixture (104) is obtained by converting it into the coordinate system of the multi-axis robot (102) through a third-party sensor for positioning; or, it is obtained by parsing the feature coordinate information of the hand-eye calibration fixture (104).

6. The calibration method according to claim 4, characterized in that, In the step where the multi-axis robot (102) moves to the target acquisition and measurement position of the vision sensor (103) according to the path planned by the information processing computer (101), the transition path executed by the multi-axis robot (102) to reach the target acquisition and measurement position is calibrated.

7. The calibration method according to claim 4, characterized in that, The steps of the information processing computer (101) acquiring point cloud data information collected by the vision sensor (103) and obtaining the feature coordinates of the hand-eye calibration fixture (104) in the coordinate system of the vision sensor (103) and the acquired pose in the coordinate system of the multi-axis robot (102) include: The filtering method is used to remove interference noise in the point cloud data information collected by the visual sensor (103); Using the structured representation and semantic segmentation method of point cloud information, the structural relationship and representation network containing only the hand-eye calibration fixture (104) features to be identified are extracted from the point cloud data information after removing interference noise points; Based on the structural relationship and representation network of the hand-eye calibration fixture (104) containing only the features to be identified, the model parameters of the measured hand-eye calibration fixture (104) are generated, and the position and attitude information of the hand-eye calibration fixture (104) in the coordinate system of the visual sensor (103) are obtained based on the model parameters of the measured hand-eye calibration fixture (104). The pose information of the hand-eye calibration fixture (104) in the coordinate system of the vision sensor (103) and the pose information of the current multi-axis robot (102) are recorded for calibration calculation.

8. The calibration method according to claim 7, characterized in that, The calibration calculation steps include establishing the relationship between corresponding points in the three-dimensional space in the coordinate system of the vision sensor (103) and the coordinate system of the multi-axis robot (102), as shown in the following expression: In the formula, The attitude of the multi-axis robot (102) is acquired in the coordinate system. The attitude of the hand-eye calibration fixture (104) under the corresponding acquired attitude of the multi-axis robot (102) is determined. Let be the attitude transformation matrix between the coordinate system of the multi-axis robot (102) and the vision sensor (103) to be solved; To acquire positions for the multi-axis robot (102), To determine the position of the hand-eye calibration fixture (104) at the corresponding acquisition position of the multi-axis robot (102), Let be the position transformation matrix between the coordinate systems of the multi-axis robot (102) and the vision sensor (103) to be solved; Using the pose information of the hand-eye calibration fixture (104) in the coordinate system of the vision sensor (103) and the pose information collected in the current coordinate system of the multi-axis robot (102), the pose transformation matrix between the coordinate system of the vision sensor (103) and the coordinate system of the multi-axis robot (102) is obtained by using the least squares method.

9. The calibration method according to claim 4, characterized in that, When the information processing computer (101) calculates the calibration relationship between the multi-axis robot (102) and the vision sensor (103) by solving the solution set of the number of feature coordinates obtained through calibration calculation and the number of coordinates of the multi-axis robot (102), it obtains the rotation relationship of the coordinate system of the vision sensor (103) relative to the coordinate system of the multi-axis robot (102). Peaceful Relationship This allows us to obtain the data transformation relationship between the two coordinate systems.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, which is executed by a processor in an electronic device to implement the calibration method as described in any one of claims 4 to 9.

Images