Segmented motion planning method and system for large-diameter penetration piece welding robot

Abstract

The invention discloses a segmented motion planning method and system for a large-diameter penetration piece welding robot. The method comprises the steps that a kinematics model of the welding robot is established, and kinematics analysis is conducted on the kinematics model; establishing a large-diameter penetration piece welding seam equation, solving main normal vectors of all discrete points on the equation, and obtaining a homogeneous transformation matrix of the tail end of the welding robot on all the discrete points; the pose matrix of the tail end of the large-diameter penetration piece welding robot of each welding section in the tail end pose matrix of the large-diameter penetration piece welding robot in the whole welding seam welding process is obtained; and inverse solution is conducted on the pose matrix of the tail end of the large-diameter penetration piece welding robot of each welding section, the corresponding angle and position of each joint of the large-diameter penetration piece welding robot at each discrete point are obtained, and trajectory planning is conducted. According to the method, the movement of the welding robot in the automatic welding process of the large-diameter penetration piece can be precisely planned, and the failure problem after welding can be reduced to the maximum extent.

Description

Technical Field

[0001] The invention relates to robots, and in particular to a segmented motion planning method and system for a large-caliber through-piece welding robot. Background Art

[0002] Welding is an important part of the manufacturing of large-caliber pressure vessels. Due to the presence of high temperature, dust, arc radiation and other welding factors, the welding working environment is often harsh. Therefore, intelligent robots have begun to enter the welding field. Using welding robots to carry out automated welding can not only improve the consistency of welding quality, but also improve production efficiency and reduce costs. It is the key to promoting intelligent manufacturing of welding work.

[0003] Large-diameter pressure vessels are common in the fields of shipbuilding and offshore engineering. Due to their special working nature, they are often in harsh service environments, and their weld surfaces are inevitably subjected to repeated effects of highly corrosive environmental media such as gas and liquid and alternating loads such as high temperature and high pressure, which undoubtedly puts higher requirements on the welding quality of intersecting line structures. Accurate motion planning is essential for automated welding, which can minimize post-welding failure problems and ensure the long-term reliability and stability of welding.

[0004] Current research mainly focuses on the welding of small-diameter through-pieces. The motion planning for welding regular or continuous welds is relatively mature. The Chinese invention patent with authorization number CN 113942017 B proposes an automatic generation method for the welding gun posture of variable curvature section welds. This method is combined with a positioner, which involves the rotation of the workpiece by the positioner during the welding process, and the weld is not segmented, which is not suitable for the welding of medium and large components of ships.

[0005] Compared with the welding of small-diameter penetrations and the welding of regular or continuous welds, the welding of large-diameter penetrations (riding and horizontal insertion) has many technical difficulties. First, due to the large size and complex structure, the welding path changes frequently, and the welding posture changes greatly and is difficult to control; then, compared with the riding penetration, the welding conditions of the horizontal insertion penetration are more complicated, and different welding processes need to be used in different welding sections, such as vertical welding, overhead welding and horizontal welding; secondly, the variable path increases the risk of equipment interference and affects the smoothness of the welding process; finally, the high-precision requirements put forward strict stability requirements for the speed and acceleration control of the end effector. Therefore, the current welding of large-diameter pressure vessels is mainly manual welding, and the welding accuracy is greatly affected by the professional and technical level of the welders, resulting in a time-consuming and labor-intensive welding process and difficult to ensure quality, which cannot meet the current industrial production processing requirements.

[0006] Therefore, research on the segmented motion planning method of welding robots for large-diameter through-pieces is carried out. The motion planning of welding robots is extremely important in the automated welding process of large-diameter through-pieces. Accurate motion planning can minimize the occurrence of post-welding failure problems and ensure the long-term reliability and stability of welding. Summary of the invention

[0007] Purpose of the invention: The purpose of the present invention is to provide a segmented motion planning method, system, equipment and storage medium for a large-diameter through-piece welding robot, which can accurately plan the movement of the welding robot during the automated welding of large-diameter through-pieces and minimize the occurrence of post-welding failure problems.

[0008] Technical solution: A segmented motion planning method for a large-caliber through-hole welding robot of the present invention comprises:

[0009] Model construction and analysis: The kinematic model of the welding robot is established using MDH parameters. Based on the kinematic model of the welding robot, the homogeneous transformation matrix between adjacent links of the large-diameter through-piece welding robot is obtained through the homogeneous transformation matrix formula. The homogeneous transformation matrix between adjacent links is right-multiplied in sequence to obtain the total transformation matrix of the coordinate system of the end of the large-diameter through-piece welding robot relative to the base coordinate system, thereby completing the kinematic analysis.

[0010] Obtaining homogeneous transformation matrix: Establishing the weld equation of large-diameter through-pieces, and solving the principal normal vector of each discrete point on the weld equation of large-diameter through-pieces; rotating the principal normal vector of each discrete point and combining it with the corresponding discrete point coordinates, obtaining the homogeneous transformation matrix of the end of the large-diameter through-piece welding robot at each discrete point, and finally obtaining the end position matrix of the large-diameter through-piece welding robot during the entire weld welding process;

[0011] Homogeneous transformation matrix acquisition of welding segments: Based on the homogeneous transformation matrix acquisition method, the pose matrix of the end position matrix of the large-diameter through-piece welding robot in each welding segment in the entire weld welding process is obtained;

[0012] Trajectory planning: The total transformation matrix obtained by kinematic analysis of the welding robot is used to inversely solve the pose matrix of the end of each welding segment large-diameter through-piece welding robot, and the angles and positions of each joint of the large-diameter through-piece welding robot at each discrete point are obtained. Then, trajectory planning is completed through the global trend adaptive difference algorithm and the reverse trajectory planning algorithm.

[0013] Furthermore, the weld equation of large-diameter penetration is established, and the principal normal vectors of each discrete point on the weld equation of large-diameter penetration are solved, including:

[0014] Simplify the actual welding object, establish a typical intersecting pipeline structure according to the actual postures of the two welded parts in the actual welding object, and establish a coordinate system corresponding to the simplified welding object based on this;

[0015] The large-diameter penetration weld equation is established based on the established coordinate system, and the large-diameter penetration weld equation is discretized. The formula of the large-diameter penetration weld equation after discretization is as follows:

[0016]

[0017] Where: R is the outer diameter of the main cylinder; r is the outer diameter of the supporting cylinder; ω is a discrete parameter variable; Δx, Δy, and Δz respectively represent the offset of the coordinate system corresponding to the simplified welding object relative to the base coordinate system of the large-diameter through-piece welding robot.

[0018] The first-order derivative of the weld curve on each discrete parameter variable ω is obtained to obtain the tangent vector T(ω) of each point, and the unit tangent vector is obtained by normalization. The specific formula is as follows:

[0019]

[0020] The resulting unit tangent vector Derivative again and normalize to obtain the principal normal vector N(ω) of each discrete point.

[0021] Furthermore, the principal normal vector N(ω) of each discrete point is calculated using the following formula:

[0022]

[0023] Furthermore, the principal normal vector of each discrete point is rotated and combined with the corresponding discrete point coordinates to obtain the homogeneous transformation matrix of the end of the large-diameter through-piece welding robot at each discrete point, and finally the end position matrix of the large-diameter through-piece welding robot during the entire weld welding process is obtained, including:

[0024] Each discrete point has a unit tangent vector and a principal normal vector. The Rodriguez rotation equation is used to rotate the principal normal vector corresponding to the current discrete point with the unit tangent vector corresponding to the current discrete point as the rotation axis to complete the rotation of the principal normal vector of each discrete point. At this time, the rotation matrix R is expressed as:

[0025]

[0026] In the formula, u x =T x / ∥T∥ represents the unit component of the tangent vector T on the X-axis; u y =T y / ∥T∥ represents the unit component of the tangent vector T on the Y axis; u z =T z / ∥T∥ represents the unit component of the tangent vector T on the Z axis; θ represents the rotation angle;

[0027] The principal normal vector N(ω) of each discrete point is rotated to obtain the rotated vector N′:

[0028] N′=RN

[0029] The coordinates (x, y, z) corresponding to each discrete point and the obtained rotated vector N′ are combined to obtain the terminal pose matrix A of the large-diameter through-piece welding robot corresponding to the discrete point:

[0030]

[0031] Where position represents the x, y, and z coordinates of the current discrete point; axis Represents the calculated rotated vector N′ as the z of the pose matrix axis ;x axis Indicates the calculated unit tangent vector as the x of the pose matrix axis ;y axis Indicates the calculation of z axis With x axis The cross product of axis .

[0032] Furthermore, based on the homogeneous transformation matrix acquisition method, the pose matrix of the end of the large-diameter through-piece welding robot in each welding segment in the end pose matrix of the large-diameter through-piece welding robot during the entire weld welding process is obtained, including:

[0033] Based on the homogeneous transformation matrix acquisition method, the pose matrix of the robot end of large-diameter through-hole welding corresponding to each discrete point on the entire weld trajectory is solved;

[0034] By specifying the segmentation form, the entire weld is divided into regions;

[0035] According to the results of area division, the pose matrix of the end of the large-diameter through-hole welding robot corresponding to each welding section is extracted.

[0036] Furthermore, the design process of the global trend adaptive difference algorithm is as follows:

[0037] Calculate the global trend factor and use it to calculate the speed of each joint of the large-diameter penetration welding robot, as follows:

[0038]

[0039] In the formula, v i,jrepresents the velocity of the jth joint at the ith discrete point; γ represents the trend factor adjustment coefficient; t i,j The time interval between adjacent discrete points;

[0040] Calculate the acceleration of each joint of the large-diameter through-piece welding robot as follows:

[0041]

[0042] a i,j represents the acceleration of the jth joint at the ith discrete point.

[0043] Furthermore, the calculation formula of the global trend factor is as follows:

[0044]

[0045] Where: S represents the global trend factor, which is used to measure the overall trend of trajectory changes; q i,j represents the position of the jth joint of the large-diameter through-hole welding robot at the i-th discrete point; △ represents the average value of the time interval;

[0046] Furthermore, the total transformation matrix obtained by kinematic analysis of the welding robot is used to inversely solve the pose matrix of the end of each welding segment large-diameter through-piece welding robot, and the angles and positions of each joint of the large-diameter through-piece welding robot at each discrete point are obtained, and then the trajectory planning is completed through the global trend adaptive difference algorithm and the reverse trajectory planning algorithm, including:

[0047] According to the actual working conditions, the large-diameter through-weld weld is discretized to obtain m nodes. For each pair of adjacent discrete points P i and P i+1 , and its corresponding welding distance Δl is calculated by Euclidean distance formula;

[0048] According to the actual working conditions, the speed v of the end of the large-diameter through-piece welding robot during uniform motion is given during the welding process, and the total welding time T is solved according to the obtained total running length and the given speed;

[0049]

[0050] The running time between each discrete point is distributed in proportion to the ratio of the distance between the segment and the total welding length to generate the corresponding time vector. The specific design formula is as follows:

[0051]

[0052] t=[t part (1),t part(2),...,t part (i)],i∈{1,2,...,m-1}

[0053] Where, t part Indicates the time between the current discrete points; T indicates the total welding time; Δl i represents the distance between adjacent discrete points, and L represents the total weld length;

[0054] The reverse trajectory planning algorithm is used to perform inverse solution on the joints of the large-diameter through-piece welding robot according to the posture matrix of the end of each welding segment large-diameter through-piece welding robot, and the angle and position data corresponding to each joint of the large-diameter through-piece welding robot at each discrete point are obtained.

[0055] According to the time vector t and the angle and position data corresponding to each joint of the large-diameter through-piece welding robot at each discrete point, the speed and acceleration of each joint during operation are solved based on the global trend adaptive difference algorithm.

[0056] Furthermore, the calculation formula of the welding distance Δl is as follows:

[0057]

[0058]

[0059] In the formula, x i ,y i ,z i They represent the coordinate values ​​of discrete points respectively.

[0060] Based on the same inventive concept, a segmented motion planning system for a large-caliber through-hole welding robot of the present invention comprises:

[0061] The model building and analysis module is used to establish the kinematic model of the welding robot using MDH parameters; based on the kinematic model of the welding robot, the homogeneous transformation matrix between adjacent links of the large-diameter through-piece welding robot is obtained through the homogeneous transformation matrix formula; the homogeneous transformation matrix between adjacent links is right-multiplied in sequence to obtain the total transformation matrix of the coordinate system of the end of the large-diameter through-piece welding robot relative to the base coordinate system, thereby completing the kinematic analysis;

[0062] The homogeneous transformation matrix acquisition module is used to establish the large-diameter through-piece weld equation and solve the principal normal vector of each discrete point on the large-diameter through-piece weld equation; the principal normal vector of each discrete point is rotated and combined with the corresponding discrete point coordinates to obtain the homogeneous transformation matrix of the large-diameter through-piece welding robot end at each discrete point, and finally obtain the end posture matrix of the large-diameter through-piece welding robot during the entire weld welding process;

[0063] A welding segment homogeneous transformation matrix acquisition module is used to obtain the posture matrix of the end of the large-diameter penetration welding robot in each welding segment in the end posture matrix of the large-diameter penetration welding robot during the entire weld welding process based on the homogeneous transformation matrix acquisition method;

[0064] The trajectory planning module is used to use the total transformation matrix obtained from the kinematic analysis of the welding robot to inversely solve the posture matrix of the end of each welding segment large-diameter through-piece welding robot, obtain the angles and positions of each joint of the large-diameter through-piece welding robot at each discrete point, and then complete the trajectory planning through the global trend adaptive difference algorithm and the reverse trajectory planning algorithm.

[0065] Beneficial effects: Compared with the prior art, the significant technical effects of the present invention are as follows:

[0066] The present invention firstly establishes a kinematic model of a large-diameter through-piece welding robot and performs kinematic analysis; then, based on the rotation vector of the principal normal vector of each discrete point of the weld, obtains the homogeneous transformation matrix of the terminal end of the large-diameter through-piece welding robot; then, this method is used to obtain the terminal posture data of the large-diameter through-piece welding robot during the entire weld welding process; then, the terminal posture data of the large-diameter through-piece welding robot of each welding section is extracted, and this is used as a benchmark to achieve the smoothing of the posture transition between sections of the large-diameter through-piece welding robot; finally, the velocity acceleration of each joint during operation is solved by a global trend adaptive differential algorithm, and the segmented motion planning of the entire welding process is completed.

[0067] The segmented posture planning method of the welding robot of the present invention is helpful to realize the real-time control of the terminal posture of the welding robot. The segmented processing can balance the heat input, avoid stress concentration and the generation of weld defects; at the same time, segmentation can be more convenient to combine with multi-posture processes such as vertical welding and overhead welding; in addition, segmented welding will also facilitate obstacle avoidance planning, effectively prevent equipment interference, and ensure the smooth progress of the welding process. In this way, the stability of the welding process of the welding robot is improved, and the purpose of optimizing welding quality and improving production efficiency is achieved.

[0068] In the stage of acquiring the pose matrix of the welding robot terminal, the weld trajectory is discretized, the main normal vector of each discrete point is rotated using the Rodriguez rotation formula, and then the homogeneous transformation matrix of the welding robot terminal is solved jointly in combination with the current discrete point coordinates. This method has the following advantages: first, the discrete processing makes the algorithm more flexible and adaptable in complex curve welding; second, through the Rodriguez rotation formula, the welding angle can be efficiently adjusted according to different welding processes, so that the welding robot terminal can meet the multi-pose processes such as vertical welding and overhead welding required for large-diameter through-pieces; finally, the homogeneous transformation matrix is ​​obtained by jointly solving, which can realize the precise adjustment of the terminal posture and ensure the continuity and stability of the welding process.

[0069] In the pose matrix extraction stage of the end of the large-diameter through-piece welding robot for each welding segment, the welding area is first divided according to the actual working conditions, and then the pose matrix corresponding to each discrete point of the entire weld trajectory is solved according to the actual welding process requirements required by different areas, and finally the end pose matrix of each welding segment is extracted. This method enables the large-diameter through-piece welding robot to accurately adjust its posture according to the complexity of different welding areas during the welding process, ensuring stable welding quality and consistent weld morphology in each welding segment. At the same time, by extracting the segmented pose data, the transition between segments can be smoothed without mutation. This method improves the flexibility and adaptability of the posture control of the large-diameter through-piece welding robot, and is suitable for segmented welding tasks with a variety of complex welding paths, thereby improving the overall welding efficiency and accuracy.

[0070] A trajectory planning method for a large-diameter through-piece welding robot based on a global trend adaptive differential algorithm is used to calculate the speed and acceleration of each joint of the large-diameter through-piece welding robot. Compared with small-diameter through-pieces, the weld is shorter and does not need to be segmented. The entire weld is welded at one time, and the speed control is relatively simple, only the initial and final speeds need to be controlled. However, large-diameter through-pieces are affected by segmentation, and the speed and acceleration at the junction of different segments still need to be accurately controlled. This method controls the uniform motion of the end effector and reversely derives the speed and acceleration of each joint at the current time node, which has the characteristics of simple and efficient algorithm. Compared with the traditional high-order polynomial trajectory planning method, this adaptive algorithm can maintain numerical stability in the case of uneven time intervals, and effectively avoid the oscillation problem caused by polynomial fitting. At the same time, this method is easier to meet the strict requirements of the end effector motion stability of large-diameter through-piece welding robots in complex welding tasks, ensuring the continuity and accuracy of the welding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a flow chart of a segmented motion planning method for a large-caliber through-hole welding robot disclosed in an embodiment of the present invention;

[0072] Figure 2 It is a flow chart of obtaining a posture matrix of a terminal end of a large-caliber through-piece welding robot disclosed in an embodiment of the present invention;

[0073] Figure 3 is a flow chart of a global trend adaptive difference algorithm disclosed in an embodiment of the present invention;

[0074] Figure 4 It is a simplified model schematic diagram of a welding scene disclosed in an embodiment of the present invention;

[0075] Figure 5This is a schematic diagram of a connecting rod coordinate system of a large-caliber penetration welding robot disclosed in an embodiment of the present invention;

[0076] Figure 6 It is a schematic diagram of a coordinate system of a part to be welded disclosed in an embodiment of the present invention;

[0077] Figure 7 It is a schematic diagram of a weld principal normal vector and its rotation disclosed in an embodiment of the present invention;

[0078] Figure 8 It is a weld segmentation and welding process diagram disclosed in an embodiment of the present invention;

[0079] Fig. 9 This is a motion trajectory diagram of a large-caliber penetration welding robot disclosed in an embodiment of the present invention;

[0080] Fig.10 It is a schematic diagram of displacement, velocity and acceleration of a robot end disclosed in an embodiment of the present invention;

[0081] Fig.11 This is a motion analysis diagram of each joint of a large-caliber penetration welding robot disclosed in an embodiment of the present invention;

[0082] Fig.12 This is a schematic diagram of the simulation of the terminal position and posture of a large-caliber through-piece welding robot disclosed in an embodiment of the present invention;

[0083] Fig.13 is a comparison result of three trajectory planning methods disclosed in the embodiments of the present invention;

[0084] Fig.14 The present invention discloses a structural diagram of a segmented motion planning system for a large-caliber through-piece welding robot. DETAILED DESCRIPTION

[0085] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It will be understood by those skilled in the art that the purposes and advantages that can be achieved with the present invention are not limited to the specific description of the above beneficial effects, and the above and other purposes that can be achieved by the present invention will be more clearly understood based on the following detailed description.

[0086] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed in the present invention can be implemented in hardware, software or a combination of the two. Whether it is implemented in hardware or software depends on the specific application and design of the technical solution. Professional and technical personnel can use different methods to implement the functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0087] Reference to "embodiments" in the present invention means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0088] Example 1

[0089] See also Figure 1 , Figure 1 The present invention discloses a flow chart of a segmented motion planning method for a large-caliber through-piece welding robot. Figure 1 The segmented motion planning method described is applied to robots, such as for planning the welding path of robots in the manufacturing process of large-caliber pressure vessels. The embodiments of the present invention are not limited. This embodiment takes the welding of large-caliber horizontal insertion pieces as an example. Figure 1 As shown, the segmented motion planning method of the large-caliber penetration welding robot may include the following operations:

[0090] S1. Model construction and analysis: Use MDH parameters to establish the kinematic model of the welding robot; based on the kinematic model of the welding robot, the homogeneous transformation matrix between the adjacent links of the large-diameter through-piece welding robot is obtained through the homogeneous transformation matrix formula; the homogeneous transformation matrix between the adjacent links is right-multiplied in sequence to obtain the total transformation matrix of the coordinate system of the end of the large-diameter through-piece welding robot relative to the base coordinate system, and then the kinematic analysis is completed.

[0091] like Figure 4 and Figure 5 As shown in Figure 1, in the model construction and analysis stage, the MDH parameters are first used to perform kinematic modeling on the large-diameter through-piece welding robot, and then the MDH parameter table is obtained, as shown in Table 1. Then, the homogeneous transformation matrix between adjacent links is obtained through the general homogeneous transformation matrix formula. The homogeneous transformation matrices between adjacent links are right-multiplied in sequence to obtain the total transformation matrix of the large-diameter through-piece welding robot end effector coordinate system relative to the base coordinate system, and then the kinematic analysis is completed.

[0092] Table 1:

[0093]

[0094] S2. Obtaining the homogeneous transformation matrix: Establish the large-diameter through-piece weld equation, and solve the principal normal vector of each discrete point on the large-diameter through-piece weld equation; rotate the principal normal vector of each discrete point and combine it with the corresponding discrete point coordinates to obtain the homogeneous transformation matrix of the large-diameter through-piece welding robot end at each discrete point, and finally obtain the end posture matrix of the large-diameter through-piece welding robot during the entire weld welding process.

[0095] like Figure 2 , Figure 6 and Figure 7 As shown, the homogeneous transformation matrix acquisition includes obtaining the principal normal vector of each discrete point of the large-diameter through-weld weld equation, and obtaining the terminal pose matrix based on the rotation of the principal normal vector. The specific steps of step S2 are as follows:

[0096] S2.1. Obtain the principal normal vectors of each discrete point in the weld equation of a large-diameter through-hole.

[0097] In this embodiment, a large-diameter penetration weld equation is established, and the principal normal vector of each discrete point on the large-diameter penetration weld equation is solved, including:

[0098] S2.1.1. Simplify the actual welding object, establish a typical intersecting pipe structure according to the actual posture of the two welded parts in the actual welding object, and establish a coordinate system corresponding to the simplified welding object based on this, which is used to calculate the weld equation, such as Figure 6 shown. Figure 6 Figure (a) is the left view of the part to be welded. Figure 6 Figure (b) shows the offset diagram between the coordinate system of the workpiece to be welded and the base coordinate system of the large-diameter through-piece welding robot.

[0099] S2.1.2. Based on the coordinate system established in step S2.1.1, establish the large-diameter penetration weld equation, discretize the large-diameter penetration weld equation, and obtain the large-diameter penetration weld equation after discretization. The details are as follows:

[0100] First, let the main cylinder rotating around the Z axis be named cylinder 1: 2 +y 2 =R 2 ; The cylinder rotating around the X axis is named cylinder 2: y 2 +z 2 =r 2 Therefore, the equation of the weld relative to the simplified weld object can be derived from this. By introducing the parameter variable ω, the equation of cylinder 2 is first parameterized, and then the discretization result is substituted into the equation x of cylinder 1. 2 +y 2 =R 2The discrete equation of the weld relative to the coordinate system corresponding to the simplified welding object can be obtained. By offsetting the discretized equation, the discretized equation of the weld relative to the base coordinate system of the large-diameter through-piece welding robot can be obtained. After completion, the formula is as follows:

[0101]

[0102] Where R is the outer diameter of the main cylinder; r is the outer diameter of the supporting cylinder; ω is a discrete parameter variable; Δx, Δy, and Δz respectively represent the offset of the coordinate system corresponding to the simplified welding object relative to the base coordinate system of the large-diameter through-piece welding robot.

[0103] S2.1.3. Take the first-order derivative of the weld curve on each discrete parameter variable ω to obtain the tangent vector T(ω) at each point, and perform normalization to obtain the unit tangent vector The specific formula is as follows:

[0104]

[0105] S2.1.4. For the unit tangent vector obtained in step S2.1.3 Derived again and normalized, we get the principal normal vector N(ω) of each discrete point, using the following formula:

[0106]

[0107] S2.2. Obtain the terminal pose matrix based on the rotation of the principal normal vector.

[0108] In this embodiment, the principal normal vector of each discrete point is rotated and combined with the corresponding discrete point coordinates to obtain the homogeneous transformation matrix of the end of the large-diameter through-piece welding robot at each discrete point, and finally the end position matrix of the large-diameter through-piece welding robot during the entire weld welding process is obtained, such as Figure 7 As shown, including:

[0109] S2.2.1. Each discrete point has a unit tangent vector and a principal normal vector. Use the Rodriguez rotation equation to rotate the principal normal vectors of each discrete point obtained in step S2.1.4 one by one with the unit tangent vector of each discrete point obtained in step S2.1.3 as the rotation axis. The rotation matrix R is expressed as:

[0110] R=Icosθ+(1-cosθ)uu T +sinθK

[0111] In the formula, u x =T x / ∥T∥ represents the unit component of the tangent vector T on the X-axis; u y =T y / ∥T∥ represents the unit component of the tangent vector T on the Y axis; u z =T z / ∥T∥ represents the unit component of the tangent vector T on the Z axis; θ represents the rotation angle;

[0112]

[0113] Substituting the above parameters into the obtained rotation matrix R is expressed as:

[0114]

[0115] In the formula, u x =T x / ∥T∥ represents the unit component of the tangent vector T on the X-axis; u y =T y / ∥T∥ represents the unit component of the tangent vector T on the Y axis; u z =T z / ∥T∥ represents the unit component of the tangent vector T on the Z axis; θ represents the rotation angle.

[0116] S2.2.2. The principal normal vector N(ω) of each discrete point is rotated to obtain the rotated vector N′, which is expressed as:

[0117] N′=RN

[0118] S2.2.3. Combine the coordinates (x, y, z) corresponding to each discrete point and the rotated vector N′ obtained by the above method to obtain the terminal pose matrix A of the large-caliber through-piece welding robot corresponding to the discrete point:

[0119]

[0120] Where position represents the x, y, and z coordinates of the current discrete point; axis Represents the calculated rotated vector N′ as the z of the pose matrix axis ;x axis Indicates the calculated unit tangent vector as the x of the pose matrix axis ;y axis Indicates the calculation of z axis With x axis The cross product of axis .

[0121] The above method is used to obtain the homogeneous transformation matrix of the end of the large-diameter through-piece welding robot, which has the following advantages. First, the algorithm has higher flexibility and adaptability in complex curve welding through discrete processing; second, the welding angle can be adjusted efficiently through the Rodriguez rotation formula; finally, the homogeneous transformation matrix is ​​obtained through simultaneous solution, which can realize the precise adjustment of the end posture and ensure the continuity and stability of the welding process.

[0122] S3. Obtaining the homogeneous transformation matrix of the welding segment: Based on the homogeneous transformation matrix acquisition method, the pose matrix of the end position matrix of the large-diameter through-piece welding robot in each welding segment in the entire weld welding process is obtained. Figure 8 and 9 As shown, the details are as follows:

[0123] S3.1. Based on the homogeneous transformation matrix acquisition method of step S2, the pose matrix of the end of the large-diameter through-hole welding robot corresponding to each discrete point on the entire weld trajectory is solved;

[0124] S3.2. By specifying the segmentation form, the entire weld can be divided into regions. Figure 8 The segmented form shown, Figure 8 Figure (a) is a schematic diagram of a horizontally inserted large-diameter penetration. Figure 8 Figure (b) is a schematic diagram of a riding-type large-diameter penetration.

[0125] When performing segmented welding on large-diameter penetrations, dual-robot symmetrical welding is generally used to weld the welds. Therefore, the overall weld is regarded as two symmetrical parts, and only half of the weld is analyzed.

[0126] S3.3. Based on the area division results, the posture matrix of the end of the large-diameter through-piece welding robot corresponding to each welding segment is extracted.

[0127] For the removal trajectory, in order to ensure the consistency of the overall calculation, the same posture planning method as the weld trajectory is selected for planning.

[0128] The pose data of the weld trajectory area is integrated with the pose data of the removal trajectory area, and the final motion trajectory of the large-diameter through-hole welding robot is as follows: Fig. 9 shown.

[0129] The method of extracting the terminal posture data of each welding segment robot adopted by the present invention enables the welding robot to make precise posture adjustments according to the complexity of different welding areas during the welding process, ensuring stable welding quality and consistent weld morphology in each welding segment. At the same time, by extracting the segmented posture data, the transition between segments can be smoothed without sudden changes. This method improves the flexibility and adaptability of the robot's posture control, is suitable for segmented welding tasks with a variety of complex welding paths, and improves the overall welding efficiency and accuracy.

[0130] S4, trajectory planning: Figure 3 , Fig.10 and Fig.11 As shown in the figure, the total transformation matrix obtained by the kinematic analysis of the welding robot is used to inversely solve the pose matrix of the end of each welding segment large-diameter through-piece welding robot, and the angle and position corresponding to each joint of the large-diameter through-piece welding robot at each discrete point are obtained, and then the trajectory planning is completed through the global trend adaptive difference algorithm and the reverse trajectory planning algorithm. The details are as follows:

[0131] S4.1. According to the actual working conditions, the large-diameter through-weld weld is discretized to obtain m nodes. For each pair of adjacent discrete points P i and P i+1 , and the corresponding welding distance Δl is calculated by the Euclidean distance formula, specifically using the following formula:

[0132]

[0133] In the formula, x i ,y i ,z i They represent the coordinate values ​​of discrete points respectively.

[0134] S4.2. According to the actual working conditions, the speed v of the end of the large-diameter through-piece welding robot during uniform motion is given during the welding process, preferably but not limited to 10 mm / s, and the total welding time T is solved according to the obtained total running length and the given speed;

[0135]

[0136] S4.3. Allocate the running time between each discrete point in proportion to the distance between the distance and the total welding length, and generate the corresponding time vector. The specific design formula is as follows:

[0137]

[0138] t=[t part (1),t part (2),...,t part(i)],i∈{1,2,...,m-1}

[0139] Where, t part Indicates the time between the current discrete points; T indicates the total welding time; Δl i represents the distance between adjacent discrete points, and L represents the total weld length.

[0140] S4.4. Using the reverse trajectory planning algorithm, according to the posture matrix of the end of the large-diameter through-piece welding robot for each welding segment obtained in step S3, the inverse solution of each joint of the large-diameter through-piece welding robot is performed to obtain the angle and position data corresponding to each joint of the large-diameter through-piece welding robot at each discrete point.

[0141] S4.5. According to the time vector t obtained in step S4.3 and the angle and position data corresponding to each joint of the large-diameter through-piece welding robot at each discrete point obtained in step S4.4, the speed and acceleration of each joint during operation are solved based on the global trend adaptive differential algorithm.

[0142] According to the above steps, the end of the welding robot can finally run at a uniform speed during the welding process. The results are as follows: Fig.10 As shown, the velocity and acceleration of each joint during operation are solved as follows Fig.11 shown. Fig.11 Figure (a) is a schematic diagram of the location. Fig.11 Figure (b) is a speed diagram. Fig.11 Figure (c) is a schematic diagram of acceleration.

[0143] In this embodiment, the design process of the global trend adaptive difference algorithm is as follows:

[0144] 1) Calculate the global trend factor:

[0145]

[0146] Where: S represents the global trend factor, which is used to measure the overall trend of trajectory changes; q i,j represents the position of the jth joint of the large-diameter through-hole welding robot at the i-th discrete point; △ Represents the average value of the time interval.

[0147] 2) The speed of each joint of the large-diameter through-piece welding robot is calculated using the global trend factor, as follows:

[0148]

[0149] In the formula, v i,j represents the velocity of the jth joint at the ith discrete point; γ represents the trend factor adjustment coefficient; t i,jThe time interval between adjacent discrete points.

[0150] 3) Calculate the acceleration of each joint of the large-diameter through-piece welding robot as follows:

[0151]

[0152] a i,j represents the acceleration of the jth joint at the ith discrete point.

[0153] In order to verify the segmented posture planning method of the large-caliber through-hole welding robot of the present invention, a simulation test of segmented welding is carried out. Fig.12 As shown in the figure, the running state of the welding robot at a certain moment in the welding process is randomly intercepted. It can be seen that the end of the welding robot always performs welding operations in a specified posture relative to the weld trajectory during the welding process. The speed planning is compared with the cubic polynomial and the quintic polynomial. The speed planning results of the three methods are shown in Fig.13 shown. Fig.13 Figure (a) is a schematic diagram of the location. Fig.13 Figure (b) is a speed diagram. Fig.13 Figure (c) is a schematic diagram of acceleration. In contrast, although cubic and quintic polynomial trajectory planning can provide smoother trajectories, oscillation problems (Runge phenomenon) are prone to occur at the beginning and end of each segment, and the computational complexity is high, requiring the solution of a set of polynomial equations. The method of the present invention controls the uniform motion of the end effector and reversely derives the velocity and acceleration of each joint at the current time node, and has the characteristics of simple and efficient algorithm. Compared with traditional high-order polynomial trajectory planning methods, this adaptive algorithm can maintain numerical stability in the case of uneven time intervals, and effectively avoid oscillation problems caused by polynomial fitting.

[0154] It should be noted that the segmented motion planning method of the large-diameter through-piece welding robot of the present invention is also applicable to the welding of large-diameter riding-type through-pieces, and it is only necessary to modify the coordinate axes when establishing the coordinate system of the welding object.

[0155] Example 2

[0156] See also Fig.14 , Fig.14 The present invention discloses a large-caliber through-hole welding robot segmented motion planning system, which can plan the robot's welding path during the manufacturing process of large-caliber pressure vessels, and specifically includes:

[0157] The model building and analysis module is used to establish the kinematic model of the welding robot using MDH parameters; based on the kinematic model of the welding robot, the homogeneous transformation matrix between adjacent links of the large-diameter through-piece welding robot is obtained through the homogeneous transformation matrix formula; the homogeneous transformation matrix between adjacent links is right-multiplied in sequence to obtain the total transformation matrix of the coordinate system of the end of the large-diameter through-piece welding robot relative to the base coordinate system, thereby completing the kinematic analysis;

[0158] The homogeneous transformation matrix acquisition module is used to establish the large-diameter through-piece weld equation and solve the principal normal vector of each discrete point on the large-diameter through-piece weld equation; the principal normal vector of each discrete point is rotated and combined with the corresponding discrete point coordinates to obtain the homogeneous transformation matrix of the large-diameter through-piece welding robot end at each discrete point, and finally obtain the end posture matrix of the large-diameter through-piece welding robot during the entire weld welding process;

[0159] A welding segment homogeneous transformation matrix acquisition module is used to obtain the posture matrix of the end of the large-diameter penetration welding robot in each welding segment in the end posture matrix of the large-diameter penetration welding robot during the entire weld welding process based on the homogeneous transformation matrix acquisition method;

[0160] The trajectory planning module is used to use the total transformation matrix obtained from the kinematic analysis of the welding robot to inversely solve the posture matrix of the end of each welding segment large-diameter through-piece welding robot, obtain the angles and positions of each joint of the large-diameter through-piece welding robot at each discrete point, and then complete the trajectory planning through the global trend adaptive difference algorithm and the reverse trajectory planning algorithm.

[0161] In an optional embodiment, the segmented motion planning method of the large-diameter through-piece welding robot includes: a) using MDH parameters to perform kinematic modeling, and then using the general homogeneous transformation matrix formula to obtain the homogeneous transformation matrix between adjacent links, and using the serial robot transformation principle to obtain the total transformation matrix of the large-diameter through-piece welding robot end effector coordinate system relative to the base coordinate system, thereby completing the kinematic analysis; b) obtaining the principal normal vector of each discrete point on the large-diameter through-piece weld equation, and obtaining the end pose matrix based on the principal normal vector; c) extracting the pose matrix of the end of the large-diameter through-piece welding robot for each welding segment; d) inversely solving the pose matrix of the end of the large-diameter through-piece welding robot for each welding segment to obtain the angles and positions of each joint of the large-diameter through-piece welding robot at each discrete point, and then completing the trajectory planning through the global trend adaptive difference algorithm and the inverse trajectory planning algorithm.

Claims

1. A segmented motion planning method for a large-caliber through-hole welding robot, characterized in that: include: Model construction and analysis: Use MDH parameters to establish the kinematic model of the welding robot; Based on the kinematic model of the welding robot, the homogeneous transformation matrix between the adjacent links of the large-diameter through-piece welding robot is obtained through the homogeneous transformation matrix formula; the homogeneous transformation matrix between the adjacent links is right-multiplied in sequence to obtain the total transformation matrix of the coordinate system of the end of the large-diameter through-piece welding robot relative to the base coordinate system, thereby completing the kinematic analysis; Obtaining homogeneous transformation matrix: Establishing the weld equation of large-diameter through-pieces, and solving the principal normal vector of each discrete point on the weld equation of large-diameter through-pieces; rotating the principal normal vector of each discrete point and combining it with the corresponding discrete point coordinates, obtaining the homogeneous transformation matrix of the end of the large-diameter through-piece welding robot at each discrete point, and finally obtaining the end position matrix of the large-diameter through-piece welding robot during the entire weld welding process; Homogeneous transformation matrix acquisition of welding segments: Based on the homogeneous transformation matrix acquisition method, the pose matrix of the end of the large-diameter through-piece welding robot in each welding segment in the end pose matrix of the large-diameter through-piece welding robot during the entire weld welding process is acquired; Trajectory planning: The total transformation matrix obtained by kinematic analysis of the welding robot is used to inversely solve the pose matrix of the end of each welding segment large-diameter through-piece welding robot, and the angles and positions of each joint of the large-diameter through-piece welding robot at each discrete point are obtained. Then, trajectory planning is completed through the global trend adaptive difference algorithm and the reverse trajectory planning algorithm.

2. The segmented motion planning method for a large-caliber through-hole welding robot according to claim 1 is characterized in that: Establish the weld equation of large-diameter penetration and solve the principal normal vectors of each discrete point on the weld equation of large-diameter penetration, including: Simplify the actual welding object, establish a typical intersecting pipeline structure according to the actual postures of the two welded parts in the actual welding object, and establish a coordinate system corresponding to the simplified welding object based on this; The large-diameter penetration weld equation is established based on the established coordinate system, and the large-diameter penetration weld equation is discretized. The formula of the large-diameter penetration weld equation after discretization is as follows: Where R is the outer diameter of the main cylinder; r is the outer diameter of the supporting cylinder; ω is a discrete parameter variable; Δx, Δy, and Δz respectively represent the offset of the coordinate system corresponding to the simplified welding object relative to the base coordinate system of the large-diameter through-piece welding robot; The first-order derivative of the weld curve on each discrete parameter variable ω is obtained to obtain the tangent vector T(ω) of each point, and the unit tangent vector is obtained by normalization. The specific formula is as follows: The resulting unit tangent vector Derivative again and normalize to obtain the principal normal vector N(ω) of each discrete point.

3. The segmented motion planning method for a large-caliber through-hole welding robot according to claim 2 is characterized in that: The principal normal vector N(ω) of each discrete point is calculated using the following formula:

4. The segmented motion planning method for a large-caliber penetration welding robot according to claim 3 is characterized in that: After the principal normal vector of each discrete point is rotated, it is combined with the corresponding discrete point coordinates to obtain the homogeneous transformation matrix of the end of the large-diameter through-piece welding robot at each discrete point, and finally the end position matrix of the large-diameter through-piece welding robot during the entire weld welding process is obtained, including: Each discrete point has a unit tangent vector and a principal normal vector. The Rodriguez rotation equation is used to rotate the principal normal vector corresponding to the current discrete point with the unit tangent vector corresponding to the current discrete point as the rotation axis to complete the rotation of the principal normal vector of each discrete point. At this time, the rotation matrix R is expressed as: In the formula, u x =T x / ∥T∥ represents the unit component of the tangent vector T on the X-axis; u y =T y / ∥T∥ represents the unit component of the tangent vector T on the Y axis; u z =T z / ∥T∥ represents the unit component of the tangent vector T on the Z axis; θ represents the rotation angle; The principal normal vector N(ω) of each discrete point is rotated to obtain the rotated vector N′: N′=RN The coordinates (x, y, z) corresponding to each discrete point and the obtained rotated vector N′ are combined to obtain the terminal pose matrix A of the large-diameter through-piece welding robot corresponding to the discrete point: In the formula, position represents the x, y, and z coordinates of the current discrete point; axis Represents the calculated rotated vector N′ as the z of the pose matrix axis ;x axis Indicates the calculated unit tangent vector as the x of the pose matrix axis ;y axis Indicates the calculation of z axis With x axis The cross product of axis .

5. The segmented motion planning method for a large-caliber penetration welding robot according to claim 1, characterized in that: Based on the homogeneous transformation matrix acquisition method, the pose matrix of the end position matrix of the large-diameter through-piece welding robot in each welding segment in the entire weld welding process is obtained, including: Based on the homogeneous transformation matrix acquisition method, the pose matrix of the robot end of large-diameter through-hole welding corresponding to each discrete point on the entire weld trajectory is solved; By specifying the segmentation form, the entire weld is divided into regions; According to the results of area division, the pose matrix of the end of the large-diameter through-hole welding robot corresponding to each welding section is extracted.

6. The segmented motion planning method for a large-caliber penetration welding robot according to claim 1, characterized in that: The design process of the global trend adaptive difference algorithm is as follows: Calculate the global trend factor and use it to calculate the speed of each joint of the large-diameter through-piece welding robot, as follows: In the formula, v i,j represents the velocity of the jth joint at the ith discrete point; γ represents the trend factor adjustment coefficient; t i,j The time interval between adjacent discrete points; Calculate the acceleration of each joint of the large-diameter through-piece welding robot as follows: a i,j represents the acceleration of the jth joint at the ith discrete point.

7. The segmented motion planning method for a large-caliber penetration welding robot according to claim 6 is characterized in that: The calculation formula of the global trend factor is as follows: Where: S represents the global trend factor, which is used to measure the overall trend of trajectory changes; q i,j represents the position of the jth joint of the large-diameter through-hole welding robot at the i-th discrete point; Represents the average value of the time interval.

8. The segmented motion planning method for a large-caliber penetration welding robot according to claim 1, characterized in that: The total transformation matrix obtained by the kinematic analysis of the welding robot is used to inversely solve the pose matrix of the end of each welding segment large-diameter through-piece welding robot, and the angles and positions of each joint of the large-diameter through-piece welding robot at each discrete point are obtained. Then, the trajectory planning is completed through the global trend adaptive difference algorithm and the reverse trajectory planning algorithm, including: According to the actual working conditions, the large-diameter through-weld weld is discretized to obtain m nodes. For each pair of adjacent discrete points P i and P i+1 , and its corresponding welding distance Δl is calculated by Euclidean distance formula; According to the actual working conditions, the speed v of the end of the large-diameter through-piece welding robot during uniform motion is given during the welding process, and the total welding time T is solved according to the obtained total running length and the given speed; The running time between each discrete point is distributed in proportion to the ratio of the distance between the segment and the total welding length to generate the corresponding time vector t. The specific design formula is as follows: t=[t part (1),t part (2),...,t part (i)],i∈{1,2,...,m-1} Where, t part Indicates the time between the current discrete points; T indicates the total welding time; Δl i represents the distance between adjacent discrete points, and L represents the total weld length; The reverse trajectory planning algorithm is used to perform inverse solution on the joints of the large-diameter through-piece welding robot according to the posture matrix of the end of each welding segment large-diameter through-piece welding robot, and the angle and position data corresponding to each joint of the large-diameter through-piece welding robot at each discrete point are obtained. According to the time vector t and the angle and position data corresponding to each joint of the large-diameter through-piece welding robot at each discrete point, the speed and acceleration of each joint during operation are solved based on the global trend adaptive difference algorithm.

9. The segmented motion planning method for a large-caliber penetration welding robot according to claim 8, characterized in that: The calculation formula of the welding distance Δl is as follows: In the formula, x i ,y i ,z i They represent the coordinate values ​​of discrete points respectively.

10. A segmented motion planning system for a large-caliber through-hole welding robot, characterized in that: include: Model building and analysis module, used to establish the kinematic model of the welding robot using MDH parameters; Based on the kinematic model of the welding robot, the homogeneous transformation matrix between the adjacent links of the large-diameter through-piece welding robot is obtained through the homogeneous transformation matrix formula; the homogeneous transformation matrix between the adjacent links is right-multiplied in sequence to obtain the total transformation matrix of the coordinate system of the end of the large-diameter through-piece welding robot relative to the base coordinate system, thereby completing the kinematic analysis; The homogeneous transformation matrix acquisition module is used to establish the large-diameter through-piece weld equation and solve the principal normal vector of each discrete point on the large-diameter through-piece weld equation; the principal normal vector of each discrete point is rotated and combined with the corresponding discrete point coordinates to obtain the homogeneous transformation matrix of the large-diameter through-piece welding robot end at each discrete point, and finally obtain the end posture matrix of the large-diameter through-piece welding robot during the entire weld welding process; A welding segment homogeneous transformation matrix acquisition module is used to obtain the posture matrix of the end of the large-diameter penetration welding robot in each welding segment in the end posture matrix of the large-diameter penetration welding robot during the entire weld welding process based on a homogeneous transformation matrix acquisition method; The trajectory planning module is used to use the total transformation matrix obtained from the kinematic analysis of the welding robot to inversely solve the posture matrix of the end of each welding segment large-diameter through-piece welding robot, obtain the angles and positions of each joint of the large-diameter through-piece welding robot at each discrete point, and then complete the trajectory planning through the global trend adaptive difference algorithm and the reverse trajectory planning algorithm.

Images