Rotary automatic calibration module, system and method of robot and 3D camera

Abstract

The application provides a rotary automatic calibration module, system and method of a manipulator and a 3D camera, which are applied to the technical field of manipulator 3D vision calibration. The automatic calibration module can be installed or separated on a calibration surface of the manipulator. The calibration surface is located at an operating end of the manipulator. The calibration module comprises a calibration plate and a calibration piece. The calibration plate is provided with a plurality of parallel guide rails at the edge position. The calibration piece comprises four column bodies and calibration discs. The calibration discs are fixedly arranged at one end of the column bodies. The other end of the column body is inserted into the guide rail to perform translational movement. The calibration module is used for realizing the coordinate system calibration of the manipulator and the 3D vision camera, reducing the redundant action of the manipulator, quickly calibrating the manipulator coordinate system and the camera coordinate system, and reversely calculating the angle of the actual action of the manipulator.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot 3D vision calibration, in particular to a rotary automatic calibration module, system and method for a robot and a 3D camera. BACKGROUND

[0002] The prerequisite for laser 3D vision to guide the action of a robot is to unify the coordinate system in the laser 3D camera with the world coordinate system of the robot, so that precise guidance can be performed.

[0003] Current existing guiding technologies can be divided into two categories:

[0004] 1. Relative value guidance: i.e. teaching template method

[0005] 2. Absolute value guidance: no need for teaching, only need to read the point information given by the vision system to implement its function according to project requirements.

[0006] In the relative value guidance technology: the execution mechanism needs to perform a teaching action in advance, and the prerequisite condition of completely meeting the project process requirements is that the vision can be further implemented only when the execution mechanism is used as a template. This method has low compatibility, and the teaching time will vary depending on the different project requirements, causing different difficulties and time uncertainty. At the same time, in specific situations such as ultra-high precision situations and extremely large number of point positions, teaching cannot be implemented and will not be accepted by the customer.

[0007] Absolute value guidance technology: the coordinate system in the 3D vision camera can be unified with the coordinate system of the execution mechanism at one time without displacement of the execution mechanism, and the point position data in the image obtained by the 3D vision is the coordinate system data in the execution mechanism. This method is fixed and has good compatibility, and if modification is needed in actual project requirements, it can be directly modified in the image without the need for a second calibration.

[0008] However, the absolute value guidance calibration technology currently used for 3D vision and six-axis robots on the market has some defects as follows:

[0009] 1. To achieve ultra-high precision calibration, the entire calibration frame must be made very complex.

[0010] 2. The calibration method is relatively complicated, and needs to be imaged by the 3D camera at different positions several times, and the robot itself needs to be set to a fixed area within the camera's field of view each time to ensure the accuracy of the calculated data.

[0011] 3. The existing calibration method is based on the actual calculation expression data of the six-axis manipulator to determine, which cannot close-loop test the pose data description of the manipulator (test the accuracy of the actual pose change data of the six-axis manipulator in the camera and the system calculation description pose data of the six-axis manipulator itself) to realize the data accuracy feedback of the manipulator so as to ensure the stability of the whole system. SUMMARY

[0012] The application provides a rotary automatic calibration module, system and method of a manipulator and a 3D camera, which reduces the redundant action of the manipulator, quickly calibrates the coordinate system of the manipulator and the coordinate system of the camera, can reversely calculate the angle of the actual action of the manipulator, and then compares with the data representation of the manipulator itself to verify the accuracy data of the angle of the manipulator, quickly confirms the performance of the manipulator, assembles the needle tip after the calibration is completed to realize the external RTCP function, increases the flexibility of the manipulator, and after the high-precision calibration is completed, can realize the multiple scanning after the manipulator grasps the product and realizes the automatic splicing function, can play the maximum posture flexibility advantage of the manipulator in the scanning process, and can also maximize the pose advantage of the six-axis manipulator when the manipulator grasps the product to execute.

[0013] The application adopts the following technical means to solve the technical problems:

[0014] The rotary automatic calibration module of the manipulator and the 3D camera provided by the application is installed or separated on the calibration surface of the manipulator, the calibration surface is located at the operating end of the manipulator, and the calibration module comprises:

[0015] A calibration plate is provided with a plurality of parallel guide rails at the edge position;

[0016] The number of calibration pieces is four, which comprises a column body and a calibration disc, the calibration disc is fixedly arranged at one end of the column body, and the other end of the column body is inserted into the guide rail to move translationally.

[0017] Further, the number of guide rails is two, which are respectively located at the edge position of the two sides of the calibration plate, and two calibration pieces are arranged on each guide rail.

[0018] Further, a plurality of mounting holes are formed at the center position of the calibration plate.

[0019] The rotary automatic calibration system of the manipulator and the 3D camera provided by the application comprises a manipulator to be calibrated, a 3D vision camera and the above-mentioned calibration module, wherein,

[0020] The calibration module is installed at the operating end of the manipulator to form a first coordinate system corresponding to the end of the manipulator;

[0021] The 3D vision camera itself has a second coordinate system, which captures and collects the calibration module installed on the manipulator to obtain the first coordinate system, and performs corresponding conversion on the second coordinate system according to the first coordinate system, realizes initial calibration, and then captures and collects the remaining positions of the manipulator, and calibrates the state information of the remaining positions to the converted second coordinate system, thereby realizing the calibration process.

[0022] The rotating automatic calibration method of the manipulator and the 3D camera according to the rotating automatic calibration system is executed, and the steps include:

[0023] The 3D vision camera captures the calibration module installed at the end position of the manipulator to obtain a first picture;

[0024] The first picture is processed to obtain a first current coordinate system of the end of the manipulator, taking the center of the four calibration members as the origin, taking the connecting line of the two calibration members on the same guide rail as the X-axis, taking the connection of the two calibration members on different guide rails as the Y-axis, and taking the height of the column and the calibration disc of a calibration member as the Z-axis.

[0025] According to the first current coordinate system, the end of the manipulator is driven to rotate along the X-axis, the Y-axis and the Z-axis by a preset angle respectively, and images are captured respectively, so that the 3D vision camera obtains a second picture (X-axis), a third picture (Y-axis) and a fourth picture (Z-axis);

[0026] The second picture (X-axis), the third picture (Y-axis) and the fourth picture (Z-axis) are processed to obtain corresponding second current coordinate system, third current coordinate system and fourth current coordinate system;

[0027] The second current coordinate system is combined and compared with the first current coordinate system to determine the X-axis of the world coordinate system in the 3D vision camera; the third current coordinate system is combined and compared with the first current coordinate system to determine the Y-axis of the world coordinate system in the 3D vision camera; the fourth current coordinate system is combined and compared with the first current coordinate system to determine the Z-axis of the world coordinate system in the 3D vision camera; thereby constructing the world coordinate system corresponding to the manipulator in the 3D vision camera;

[0028] The 3D vision camera collects the remaining positions outside the end position of the manipulator and corresponds to imaging in the world coordinate system, thereby realizing the calibration process.

[0029] Further, in the step of driving the end of the manipulator to rotate along the X-axis, the Y-axis and the Z-axis by a preset angle respectively:

[0030] The preset angle is 10° to 20°.

[0031] Further, the step of obtaining the first current coordinate system of the end of the manipulator includes:

[0032] The point conversion algorithm is adopted to import the first current coordinate system into the 3D vision camera, and the point conversion algorithm is as follows:

[0033]

[0034] Among them, X, Y, and Z in the world coordinate system, and 1 is the data of the origin of the first current coordinate system of the operating end of the manipulator converted to the origin of the world coordinate system; X A , Y A , and Z A are three-axis data of the first current coordinate system, respectively; is the hand-eye conversion matrix of the manipulator.

[0035] Further, the step of combining and comparing the second current coordinate system with the first current coordinate system to determine the X-axis of the world coordinate system in the 3D vision camera, combining and comparing the third current coordinate system with the first current coordinate system to determine the Y-axis of the world coordinate system in the 3D vision camera, and combining and comparing the fourth current coordinate system with the first current coordinate system to determine the Z-axis of the world coordinate system in the 3D vision camera, thereby constructing the world coordinate system corresponding to the manipulator in the 3D vision camera, includes:

[0036] The rotation conversion algorithm is adopted to calibrate the world coordinate system, and the conversion algorithm is as follows:

[0037]

[0038] Among them, is the hand-eye conversion matrix of the manipulator, and X A1～An is the sampling X-axis point data of the rotation starting point, the process, and the rotation key point; Y A1～An is the sampling Y-axis point data of the rotation starting point, the process, and the rotation key point; and Z A1～An is the sampling Z-axis point data of the rotation starting point, the process, and the rotation key point.

[0039] Further, the step of combining and comparing the second current coordinate system with the first current coordinate system to determine the X-axis of the world coordinate system in the 3D vision camera, combining and comparing the third current coordinate system with the first current coordinate system to determine the Y-axis of the world coordinate system in the 3D vision camera, and combining and comparing the fourth current coordinate system with the first current coordinate system to determine the Z-axis of the world coordinate system in the 3D vision camera, thereby constructing the world coordinate system corresponding to the manipulator in the 3D vision camera, includes: is determined by 3D camera pre-training, and the pre-training method includes:

[0040] The 3D camera randomly positions at least three points on the manipulator;

[0041] The hand-eye conversion matrix is obtained by hand-eye conversion of the data of the three points;

[0042]

[0043] Wherein, 1~2~3 are three points on the mechanical arm randomly determined by the 3D camera.

[0044] The application provides a rotating automatic calibration module, system and method of a mechanical arm and a 3D camera, which has the following beneficial effects:

[0045] The automatic calibration module provided by the application can be installed or separated on a calibration surface of the mechanical arm, the calibration surface is located at an operating end of the mechanical arm, and the calibration surface comprises a calibration plate and a calibration piece. The calibration plate is provided with a plurality of parallel guide rails at a peripheral position. The calibration piece comprises a column body and a calibration disc, the calibration disc is fixedly arranged at one end of the column body, and the other end of the column body is inserted into the guide rail to move translationally. The calibration module is used for realizing calibration of coordinate systems of the mechanical arm and the 3D camera, reducing redundant actions of the mechanical arm, quickly calibrating the coordinate system of the mechanical arm and the coordinate system of the camera, reversely calculating an angle of actual action of the mechanical arm, comparing the angle with data of the mechanical arm to verify precision data of the angle of the mechanical arm, quickly confirming performance of the mechanical arm, realizing an external RTCP function after assembly of a needle tip after calibration, increasing flexibility of the mechanical arm, realizing multiple scanning of the mechanical arm after high-precision calibration and realizing an automatic splicing function, and maximizing flexibility of the mechanical arm in a scanning process. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 FIG. 1 is a schematic diagram of an overall structure of an embodiment of a rotating automatic calibration module of a mechanical arm and a 3D camera according to the application.

[0047] Figure 2 FIG. 2 is a schematic diagram of a flow of an embodiment of a rotating automatic calibration method of a mechanical arm and a 3D camera according to the application.

[0048] The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0049] It should be understood that the specific embodiments described herein merely serve to explain the application and are not intended to limit the application.

[0050] The technical solutions in the embodiments of the application will be described clearly and completely with reference to the accompanying drawings in the embodiments of the application. Obviously, the described embodiments are only some of the embodiments of the application, but not all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the application.

[0051] It should be noted that the terms "comprise", "comprising", "include", "including", and "have" and "having" in the specification and claims of this application, and the above-mentioned accompanying drawings, are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to such processes, methods, products or devices. In the claims, specification and drawings of this application, the terms such as "first" and "second" and the like relationship terms are only used to distinguish one entity / operation / object from another entity / operation / object, and do not necessarily require or imply any such actual relationship or order between the entities / operations / objects.

[0052] Reference herein to "embodiments" means that the particular features, structures or characteristics described in connection with the embodiments can be included in at least one embodiment of the application. The occurrence of the phrase in various places in the specification does not necessarily all refer to the same embodiment, nor is it necessarily independent or alternative to other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0053] Reference is made to the accompanying drawings Figure 1 A schematic structural view of a mechanical hand and a rotating automatic calibration module of a 3D camera in an embodiment of the present application;

[0054] The rotating automatic calibration module of the mechanical hand and the 3D camera proposed in the present application is arranged on the calibration surface of the mechanical hand and can be installed or separated, the calibration surface is located at the operating end of the mechanical hand, and the calibration module comprises:

[0055] A calibration plate 1 is provided with a plurality of parallel guide rails 2 at the edge position;

[0056] The calibration pieces 3 are four in number and comprise a column body 31 and a calibration disc 32, the calibration disc 32 is fixedly arranged at one end of the column body 31, and the other end of the column body 31 is inserted into the guide rail 2 for translational movement.

[0057] Specifically,

[0058] The above-mentioned guide rails 2 are two in number and are respectively located at the edge position on both sides of the calibration plate 1, and two calibration pieces 3 are arranged on each guide rail 2.

[0059] The above-mentioned calibration plate 1 is provided with a plurality of mounting holes 4 at the center position.

[0060] In the implementation process, the calibration module is installed on the operating end calibration surface of the manipulator through the installation hole 4, the calibration piece 3 moves on the calibration plate 1 through the guide rail 2, and the user can adjust the position of the calibration piece 3. It needs to be explained that the 3D vision camera collects the position of the calibration piece 3 to determine the first coordinate system of the manipulator, and the position of the calibration piece 3 is adjusted to adapt to the calibration of the manipulator in various forms.

[0061] In one embodiment, the present application provides a rotary automatic calibration system of a manipulator and a 3D camera, which comprises a manipulator to be calibrated, a 3D vision camera and the above-mentioned calibration module, wherein,

[0062] The calibration module is installed on the operating end of the manipulator to form a first coordinate system corresponding to the end of the manipulator;

[0063] The 3D vision camera itself has a second coordinate system, which captures and collects the calibration module installed on the manipulator to obtain the first coordinate system, and performs corresponding conversion on the second coordinate system according to the first coordinate system to realize initial calibration, and then captures and collects the remaining positions of the manipulator, and calibrates the state information of the remaining positions to the converted second coordinate system to realize the calibration process.

[0064] Reference is made to the accompanying drawings Figure 2 The flowchart of the rotary automatic calibration method of the manipulator and the 3D camera proposed in the present application.

[0065] A rotary automatic calibration method of a manipulator and a 3D camera, which performs the rotary automatic calibration method according to the above-mentioned rotary automatic calibration system, and the steps include:

[0066] S100, the 3D vision camera captures the calibration module installed on the end position of the manipulator to obtain a first picture;

[0067] S200, the first picture is processed to obtain a first current coordinate system of the end of the manipulator, taking the center of the four calibration pieces 3 as the origin, taking the connecting line of the two calibration pieces 3 on the same guide rail 2 as the X axis, taking the connection of the two calibration pieces 3 on different guide rails 2 as the Y axis, and taking the height of the column 31 and the calibration disc 32 of one calibration piece 3 as the Z axis;

[0068] S300, according to the first current coordinate system, the end of the manipulator is driven to rotate along the X axis, the Y axis and the Z axis respectively by a preset angle, and images are captured respectively, so that the 3D vision camera obtains a second picture (X axis), a third picture (Y axis) and a fourth picture (Z axis);

[0069] S400, the second picture (X axis), the third picture (Y axis) and the fourth picture (Z axis) are processed to obtain corresponding second current coordinate system, third current coordinate system and fourth current coordinate system.

[0070] S500, the second current coordinate system is combined and compared with the first current coordinate system, the X axis of the world coordinate system in the 3D vision camera is determined; the third current coordinate system is combined and compared with the first current coordinate system, the Y axis of the world coordinate system in the 3D vision camera is determined; the fourth current coordinate system is combined and compared with the first current coordinate system, the Z axis of the world coordinate system in the 3D vision camera is determined; and thus the world coordinate system corresponding to the 3D vision camera and the manipulator is constructed.

[0071] S600, the 3D vision camera collects the positions other than the end position of the manipulator, and images in the world coordinate system, so as to realize the calibration process.

[0072] In one embodiment, in the step of driving the end of the manipulator to rotate by a preset angle along the X axis, the Y axis and the Z axis:

[0073] The preset angle is 10° to 20°.

[0074] Specifically, the step of obtaining the first current coordinate system of the end of the manipulator includes:

[0075] The first current coordinate system is imported into the 3D vision camera by using a point conversion algorithm, and the point conversion algorithm is:

[0076]

[0077] Wherein, X, Y and Z in the world coordinate system are the origin data of the first current coordinate system of the end of the manipulator converted to the origin of the world coordinate system; X A , Y A and Z A are three-axis data of the first current coordinate system, respectively. is the hand-eye conversion matrix of the manipulator.

[0078] In one embodiment, the second current coordinate system is combined and compared with the first current coordinate system, the X axis of the world coordinate system in the 3D vision camera is determined; the third current coordinate system is combined and compared with the first current coordinate system, the Y axis of the world coordinate system in the 3D vision camera is determined; the fourth current coordinate system is combined and compared with the first current coordinate system, the Z axis of the world coordinate system in the 3D vision camera is determined; and thus the step of constructing the world coordinate system corresponding to the 3D vision camera and the manipulator includes:

[0079] The calibration of the world coordinate system is performed by using a rotation conversion algorithm, and the conversion algorithm is:

[0080]

[0081] wherein, H is the hand-eye transformation matrix of the robot hand, wherein X A1～An is the sampling X-axis point position data of the rotation starting point, process, and rotation key point; Y A1～An is the sampling Y-axis point position data of the rotation starting point, process, and rotation key point; Z A1～An is the sampling Z-axis point position data of the rotation starting point, process, and rotation key point.

[0082] The above is determined by 3D camera pre-training, and the pre-training method includes:

[0083] The 3D camera randomly positions at least three points on the robot hand;

[0084] The hand-eye transformation matrix is obtained by performing hand-eye transformation on the data of the three points;

[0085]

[0086] wherein, 1~2~3 are three points on the robot hand randomly determined by the 3D camera.

[0087] In summary, the automatic calibration module proposed in the present application can be installed or separated on the robot hand calibration surface, the calibration surface is located at the operating end of the robot hand, and the calibration surface includes a calibration plate 1 and a calibration piece 3, wherein the calibration plate 1 is provided with a plurality of parallel guide rails 2 at the edge position; the calibration piece 3 is four in number and includes a column body 31 and a calibration disc 32, the calibration disc 32 is fixedly arranged at one end of the column body 31, and the other end of the column body 31 is inserted into the guide rail 2 for translational movement; it is used to realize the calibration of the coordinate system of the robot hand and the 3D vision camera, to reduce the redundant action of the robot hand, to quickly calibrate the robot hand coordinate system and the camera coordinate system, to reversely calculate the angle of the actual action of the robot hand, and then to compare with the data representation of the robot hand itself to verify the accuracy data of the angle of the robot hand, to quickly confirm the performance of the robot hand, to realize the external RTCP function after the calibration is completed, to increase the flexibility of the robot hand, and to realize the automatic splicing function after the high-precision calibration is completed, and to play the maximum pose flexibility advantage of the robot hand in the scanning process, and to also maximize the pose advantage of the six-axis robot hand when the robot hand is executing the product.

[0088] For the on-site application of the calibration module proposed in the present application, the following advantages are obtained:

[0089] 1. The robot hand and 3D vision rotation automatic calibration model is simple to operate, has strong replicability, and can reversely calculate the motion accuracy of the actual robot hand, so it is suitable for all industrial six-axis robot hand operation scenes.

[0090] 2. The need for mechanical hand to grab the product to achieve true RTCP function, and the need for multiple faces of the product to be scanned and then spliced to perform all trajectories at one time.

[0091] 3. The automatic splicing function of all the words of this model realizes the substantial improvement of the accuracy of image automatic splicing. Currently, it is used in the actual cases of polishing of shoe soles and glue spraying of shoe soles. These cases all use RTCP function (mechanical hand carrying product to move and adjust posture). Since the polishing head of the polishing industry rotates at high speed, most of them are based on external fixation for operation. RTCP guiding function just realizes this 3D visual guiding function. In the dispensing and painting industry, the glue outlet of the needle will be affected by gravity and inertia, resulting in abnormal glue path of the dispensing glue. Therefore, the most ideal way for the dispensing needle is to fix the needle externally, and the mechanical hand carries the product to move and dispense. At this time, the dispensing points executed by the mechanical hand are obtained by the 3D visual system.

[0092] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A robotic hand and 3D camera rotational auto-calibration module, comprising: The calibration module can be mounted or separated from the manipulator calibration surface, the calibration surface is located at the operating end of the manipulator, and the calibration module comprises: A calibration plate is provided with a plurality of parallel guide rails at the edge position; Four calibration members, which include a column and a calibration disc, the calibration disc is fixedly arranged at one end of the column, and the other end of the column is inserted into the guide rail for translational movement; The guide rails are two, which are respectively located at the edge position on both sides of the calibration plate, and each of the guide rails is provided with two calibration members; A plurality of mounting holes are formed at the center position of the calibration plate.

2. A mechanical hand and 3D camera rotational automatic calibration system, characterized in that, The system comprises a manipulator to be calibrated, a 3D visual camera and the calibration module of claim 1, wherein The calibration module is mounted to the operating end of the manipulator to form a first coordinate system corresponding to the end of the manipulator; The 3D visual camera itself has a second coordinate system, which captures and collects the calibration module mounted on the manipulator to obtain the first coordinate system, and correspondingly converts the second coordinate system according to the first coordinate system to realize initial calibration, and then captures and collects the remaining positions of the manipulator to calibrate the state information of the remaining positions to the converted second coordinate system to realize the calibration process.

3. A method for automatic calibration of a robot and a 3D camera in rotation, characterized in that, The rotating automatic calibration system of claim 2 performs the rotating automatic calibration method, and the steps comprise: The 3D visual camera captures the calibration module mounted at the end position of the manipulator to obtain a first picture; Picture processing is performed on the first picture, the center of the four calibration members is taken as the origin, the connecting line of the two calibration members on the same guide rail is taken as the X-axis, the connection of the two calibration members on different guide rails is taken as the Y-axis, and the height of the column and the calibration disc of one calibration member is taken as the Z-axis, so as to obtain the first current coordinate system of the end of the manipulator; According to the first current coordinate system, the end of the manipulator is driven to rotate along the X-axis, the Y-axis and the Z-axis by a preset angle respectively, and images are captured respectively, so that the 3D visual camera obtains a second picture (X-axis), a third picture (Y-axis) and a fourth picture (Z-axis); Picture processing is performed on the second picture (X-axis), the third picture (Y-axis) and the fourth picture (Z-axis) to finally obtain the corresponding second current coordinate system, the third current coordinate system and the fourth current coordinate system; The second current coordinate system is combined and compared with the first current coordinate system to determine the X-axis of the world coordinate system in the 3D visual camera; the third current coordinate system is combined and compared with the first current coordinate system to determine the Y-axis of the world coordinate system in the 3D visual camera; the fourth current coordinate system is combined and compared with the first current coordinate system to determine the Z-axis of the world coordinate system in the 3D visual camera; and thus the world coordinate system corresponding to the manipulator in the 3D visual camera is constructed; The 3D visual camera collects the remaining positions outside the end position of the manipulator and correspondingly images in the world coordinate system to realize the calibration process; In the step of driving the end of the manipulator to rotate along the X-axis, the Y-axis and the Z-axis by a preset angle respectively: The preset angle is 10° to 20°.

4. The method of claim 3, wherein, The step of obtaining the first current coordinate system of the end of the manipulator comprises: The first current coordinate system is introduced into the 3D visual camera by using a point conversion algorithm, wherein the point conversion algorithm is: wherein, X, Y, Z in the equation (1) are the world coordinate system, and 1 is the origin data of the first current coordinate system of the operating end of the robot converted to the origin of the world coordinate system; X A , Y A , Z A in the equation (2) are the three-axis data of the first current coordinate system, respectively; is the hand-eye conversion matrix of the robot.

5. The method of claim 4, wherein, The second current coordinate system is combined with the first current coordinate system for comparison, so as to determine the X axis of the world coordinate system in the 3D visual camera; the third current coordinate system is combined with the first current coordinate system for comparison, so as to determine the Y axis of the world coordinate system in the 3D visual camera; the fourth current coordinate system is combined with the first current coordinate system for comparison, so as to determine the Z axis of the world coordinate system in the 3D visual camera; thereby the steps of constructing the world coordinate system corresponding to the 3D visual camera and the robot include: The world coordinate system is calibrated by using a rotation conversion algorithm, wherein the conversion algorithm is: Wherein, H is the hand-eye transformation matrix of the robot hand, wherein X A1～An X is the sampling X-axis point position data of the rotation starting point, the rotation process, and the rotation key point; A1～An Y is the sampling Y-axis point position data of the rotation starting point, the rotation process, and the rotation key point; A1～An Z is the sampling Z-axis point position data of the rotation starting point, the rotation process, and the rotation key point.

6. The method of claim 4 or 5, wherein, The For pre-training determination by a 3D camera, the pre-training manner includes: The 3D camera randomly positions at least three points on the robot; The hand-eye conversion matrix is obtained by performing hand-eye conversion on the data of the three points; Wherein, 1-2-3 are three points on the robot randomly determined by the 3D camera.

Images