Welding track planning method for four-footed humanoid welding robot

Abstract

The invention provides a welding track planning method for a four-footed humanoid welding robot. The welding track planning method comprises the steps that the welding robot is navigated to a welding station, and welding seam information is recognized and extracted; motion trail planning is conducted on the double mechanical arms in an interpolation mode, and a mechanical arm tail end trail sequence generated through interpolation is obtained; the inverse kinematics of the double mechanical arms is solved for each pose, multiple sets of joint angular solutions are obtained, and the optimal solution which is coherent in motion and minimum in deviation is screened out; the condition number of the Jacobian matrix of each point of the joint track sequence is calculated, and the potential singular pose of the mechanical arm is analyzed and recognized; needed damping factors are calculated in real time according to the singular degree, and the joint speed of the mechanical arm is smoothly adjusted when a singular point is approached; and collision avoidance analysis is carried out on the tracks of the double mechanical arms. According to the method, by combining inverse kinematics multi-solution optimization selection, trajectory time re-parameterization, a damping least square control strategy and a double-arm space coordination collision avoidance mechanism, the stability, safety and consistency in the welding task execution process are ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of autonomous welding, more particularly to a four-legged humanoid welding robot welding trajectory planning method. BACKGROUND

[0002] With the development of intelligent manufacturing technology, welding robots are widely used in automobile manufacturing, shipbuilding, aerospace, steel structure engineering and other fields. Compared with single-arm welding robots, humanoid welding robots have higher flexibility and can perform multiple welding tasks simultaneously or cooperate to complete complex welds. However, there are still many problems in the trajectory planning of humanoid welding robots, which restrict their application in high-precision welding scenarios.

[0003] Currently, the main problems of welding trajectory planning include the following aspects: first, in the process of moving from the starting position to the welding area, the traditional path planning method is usually based on geometric path or fixed path interpolation, which fails to fully consider environmental obstacles, motion interference between two arms and energy optimization problems. This planning method may result in long motion trajectory of the arm, high energy consumption, and even collision during motion, affecting welding efficiency and safety. Secondly, during the welding process, the trajectory of the welding torch along the weld determines the quality of the welding. However, existing methods often ignore the optimization of fusion zone, heat-affected zone and weld sequence during welding, resulting in unstable weld quality and even welding defects. In addition, due to the high degree of freedom of humanoid robots, the smoothness of the trajectory during motion is an important factor affecting the quality of welding. If the trajectory has problems such as sudden change of speed, sudden stop and sudden movement, it may lead to problems such as decrease in weld quality and uneven weld fusion. However, existing trajectory planning methods lack optimization for welding motion smoothness, and cannot effectively reduce the problems of trajectory shock and sudden change of speed.

[0004] Another important factor affecting welding precision is the singularity problem. When the robot is in some specific pose, the Jacobian matrix J(θ) may become non-invertible or approach a singularity, resulting in unstable inverse kinematics solution of the robot, drastic change of joint speed, and even system out-of-control. Currently, the avoidance of singularities mainly relies on manual adjustment of motion trajectory or addition of extra redundant degrees of freedom, but there are limitations in the selection of robots.

[0005] Therefore, it is necessary to propose a dual-arm welding robot trajectory planning method that integrates path planning, obstacle avoidance planning, trajectory smoothing processing and singularity avoidance, in order to improve welding efficiency, reduce welding defects, and enhance the stability and safety of dual-arm collaborative work. SUMMARY

[0006] In order to overcome the defects and deficiencies in the prior art, the purpose of the present application is to provide a quadruped humanoid welding robot welding trajectory planning method to solve the problems of insufficient welding trajectory optimization, limited obstacle avoidance capability, low synchronous control precision, lack of trajectory smoothness and insufficient singular point avoidance capability in the prior art; the method combines inverse kinematics multi-solution optimization selection, trajectory time reparameterization, damping least square control strategy and dual-arm space coordination collision avoidance mechanism to ensure the stability, safety and consistency during the execution of the welding task.

[0007] In order to achieve the above purpose, the present application is implemented by the following technical scheme: a quadruped humanoid welding robot welding trajectory planning method based on a quadruped humanoid welding robot with dual mechanical arms; the method comprises the following steps:

[0008] S1, the welding robot navigates to a welding station, identifies and extracts weld information;

[0009] S2, the dual mechanical arms are subjected to interpolation motion trajectory planning to obtain an interpolation generated mechanical arm end trajectory sequence; the motion trajectory is divided into two segments, which are a first segment motion trajectory from an initial pose to a welding start pose , and a second segment motion trajectory from the welding start pose to a welding end pose ; i=1,2, representing two mechanical arms;

[0010] S3, for each pose of the mechanical arm end trajectory sequence, inverse kinematics of the dual mechanical arms is solved to obtain multiple sets of joint angle solutions and to select the optimal solution with the most coherent motion and the smallest deviation to construct a complete joint trajectory sequence

[0011] S4, the dual mechanical arms execute the joint trajectory sequence of the first segment motion trajectory to adjust from the initial pose to the welding start pose

[0012] S5, adjustment is made for the second segment motion trajectory, including:

[0013] S51, the condition number K(J(θ)) of the Jacobian matrix J(θ) of each point of the joint trajectory sequence is calculated to analyze and identify potential mechanical arm singular poses;

[0014] S52, the required damping factor λ is calculated in real time according to the degree of singularity to smoothly adjust the mechanical arm joint speed when approaching the singular point;

[0015] S53, collision avoidance analysis is performed on the trajectories of the dual robot arms, and when potential collision is detected, the movement speed and timing of the dual robot arms are adjusted to avoid interference and ensure synchronous operation;

[0016] S6, the dual robot arms execute the joint trajectory sequence of the second segment of the movement trajectory;

[0017] S7, after the welding is completed, the dual robot arms are reset to the initial position.

[0018] Preferably, the step S51 refers to:

[0019] For the second segment of the movement trajectory, each joint trajectory point of the joint trajectory sequence By singular value decomposition of the Jacobian matrix J(θ):

[0020] J(θ) = U∑V T

[0021] ∑ = diag(σ1,σ2,…,σ n )

[0022] Wherein, U and V are orthogonal matrices; ∑ is a diagonal matrix, and the diagonal elements are singular values σ1,σ2,…,σ n .

[0023] The condition number κ(J(θ)) is calculated:

[0024]

[0025] Wherein, σ max (J(θ)) represents the maximum singular value of the Jacobian matrix, and σ min (J(θ)) represents the minimum singular value of the Jacobian matrix; a singularity judgment threshold κ max is set, if the current condition number κ(J(θ)) exceeds the singularity judgment threshold κ max , it is judged as a potential singular pose, and a deceleration, avoidance or interpolation adjustment strategy is adopted.

[0026] Preferably, the step S52 refers to: for the second segment of the movement trajectory, a damped least square method is used to solve the stability of the joint speed:

[0027]

[0028] Wherein, is the joint speed vector; is the speed vector of the end of the robot arm; λ is the damping factor, and I is the unit matrix;

[0029] The damping factor λ is dynamically adjusted according to the condition number κ(J(θ)):

[0030]

[0031] wherein λ0 is a basic damping value, κ safe is a condition number warning threshold, and δ is a growth exponent.

[0032] Preferably, the step S53 refers to: for the second segment of the motion trajectory, calculating the minimum spatial distance d i , of the two robot arms at each time t 12 ; i

[0033] When d 12 (t) < d safe at any time, d safe is a set minimum safety distance threshold, and it is determined that there is a potential interference between the two robot arms; the motion speed and timing of the two robot arms are fine-tuned to ensure that d 12 (t i ) ≥ d safe .

[0034] Preferably, the step S2 refers to: comprising the following steps:

[0035] S21, acquiring initial positions and poses of the two robot arms

[0036] The welding seam information of the step S1 includes: a welding seam path and welding poses required by the welding seam;

[0037] According to the three-dimensional coordinates of the welding seam path, three-dimensional coordinates of welding start positions and welding end positions of the two robot arms are obtained and the pose of the robot arm end in the main welding motion trajectory is fixedly limited to

[0038] The initial poses of the two robot arms are constructed the welding start pose and the welding end pose are:

[0039]

[0040] S22, for the first segment of the motion trajectory from the initial pose to the welding start pose and the second segment of the motion trajectory from the welding start pose to the welding end pose , interpolation method is used for refinement processing to calculate the position and pose of each interpolation point

[0041] ​S23, combine the position and pose of each interpolation point in the motion trajectory into a pose get the interpolation generated end-of-arm trajectory sequence:

[0042] Preferably, the step S22 refers to: respectively calculating the Euclidean distance L between the start point S and the end point T of the segment motion trajectory of the double end-of-arm (i) ; calculate the number of interpolation points required:

[0043]

[0044] wherein d is the set interpolation step; N (i) The value of N should be rounded up;

[0045] For each interpolation point j (i) =1,2,…,N (i) , calculate the position of the j (i) th interpolation point:

[0046]

[0047] wherein, is the position of the start point S; is the position of the end point T; represents the proportion where the current interpolation point is located;

[0048] For the first segment of motion trajectory, the pose of the j (i) th interpolation point is:

[0049]

[0050] wherein, is the pose quaternion of the end-of-arm at the start point S, is the pose quaternion of the end-of-arm at the end point T; is the included angle between the two quaternions; after interpolation, is converted into a rotation matrix to get the pose

[0051] For the second segment of motion trajectory, the pose of each interpolation point is limited to

[0052] Preferably, the step S3 refers to:

[0053] for each pose of the end-of-arm trajectory sequence use inverse kinematics algorithm to solve all joint angle combinations that satisfy the pose: ​​

[0054]

[0055] in, Indicates the jth (i) The nth inverse solution of the pose; k is the number of legal inverse solutions of the current pose;

[0056] Set the cost function to:

[0057]

[0058] in, Indicates the optimal inverse solution selected for the previous pose; represents the Jacobian matrix corresponding to the joint angle; κ(J) represents the condition number of the Jacobian matrix, which is used to quantify the degree of proximity to the singular point; ||·|| represents the Euclidean distance in the joint space, w1 and w2 are weight parameters; σ max (J(θ)) represents the maximum singular value of the Jacobian matrix, σ min (J(θ)) represents the minimum singular value of the Jacobian matrix;

[0059] For the jth (i) Positions Traverse the inverse solution set Θ j(i) , calculate each inverse solution The cost function Select the optimal solution

[0060]

[0061] Repeat the above process until all poses The optimal solution is determined Finally, a complete joint trajectory sequence is constructed

[0062] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0063] 1. Under the premise that the end-arm posture is fully constrained, this invention adopts an inverse kinematics multi-solution screening mechanism in the trajectory planning stage and solves the joint velocity using the damped least squares method in the trajectory execution stage. This solves the problems of joint space discontinuity and manipulator jitter caused by multiple sets of inverse solutions in traditional trajectory planning.

[0064] 2. The present invention introduces the Jacobian matrix condition number as a singularity judgment indicator for the robotic arm, which can accurately identify the numerical stability of the current robot posture;

[0065] 3. The present invention uses dual robotic arms to perform dual-gun synchronous welding on a single weld to improve welding efficiency. To address the spatial interference problem that exists in the collaborative operation of dual arms, the present invention introduces a dual-arm collision avoidance mechanism, which calculates the minimum distance between the two arms in real time and dynamically coordinates the execution speed to achieve time-staggered collision avoidance. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 1 is a schematic structural diagram of the quadruped humanoid welding robot of the present invention;

[0067] Figure 2 is a flow chart of the welding trajectory planning method of the quadruped humanoid welding robot of the present invention;

[0068] Among them, 1 represents the welding gun; 2 represents the right robotic arm; 3 represents the welding gun; 4 represents the left robotic arm; 5 represents the lidar; 6 represents the depth camera; 7 represents the quadruped robot body; 8 represents the welding power supply. DETAILED DESCRIPTION

[0069] The present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

[0070] Example

[0071] This embodiment provides a welding trajectory planning method for a quadruped humanoid welding robot, which is characterized by: based on a quadruped humanoid welding robot; Figure 1 As shown, the quadruped humanoid welding robot is provided with dual robotic arms, including a left robotic arm 4 and a right robotic arm 2; the end effectors of the robotic arms are welding guns 1 and 3.

[0072] The two manipulators in the present invention are both six-axis manipulators, that is, each manipulator has 6 degrees of freedom. Taking one of the manipulators as an example, the joint angles are defined as θ1, θ2, θ3, θ4, θ5, θ6, so the joint angle vector is θ = [θ1, θ2, ..., θ6] T , and the position of the end of the robot arm (welding gun) can be expressed by the forward kinematics equation as x = f(θ), where f(θ) is calculated by the DH parameters.

[0073] Welding trajectory planning method for quadruped humanoid welding robot, such as Figure 2 As shown, the following steps are included:

[0074] S1. The welding robot navigates to the welding station and identifies and extracts weld information.

[0075] S2, use interpolation method to plan the motion trajectory of the dual manipulators, and obtain the interpolated manipulator end trajectory sequence; the motion trajectory is divided into two sections, namely, from the initial posture To the welding starting position The first motion trajectory, and the starting position from welding to the welding termination pose , i = 1, 2, represent two mechanical arms. Specifically, the following steps are included:

[0076] S21, obtaining the initial position and pose of the end of the double mechanical arm

[0077] The welding seam information of step S1 includes: a welding seam path and welding poses required by the welding seam;

[0078] According to the three-dimensional coordinates of the welding seam path, the three-dimensional coordinates of the welding start position and the welding termination position of the end of the double mechanical arm are obtained and the pose of the end of the mechanical arm in the main welding motion trajectory is fixedly limited to

[0079] The initial pose of the end of the double mechanical arm is constructed the welding start pose and the welding termination pose are as follows:

[0080]

[0081] S22, for the first segment motion trajectory from the initial pose to the welding start pose , and the second segment motion trajectory from the welding start pose to the welding termination pose , interpolation refinement processing is performed by using an interpolation method to calculate the position and pose of each interpolation point

[0082] The Euclidean distance L between the start point S and the end point T of the segment motion trajectory of the end of the double mechanical arm is calculated (i) :

[0083]

[0084] wherein,

[0085] The number of interpolation points required is calculated:

[0086]

[0087] wherein, d is a set interpolation step, the value of which determines the smoothness of the planned path; the value of N (i) should be rounded up;

[0088] For each interpolation point j (i) = 1, 2, …, N (i) ​, calculate the jth (i) The position of the interpolation point for:

[0089]

[0090] in, is the position of the starting point S; is the position of the end point T; Indicates the scale of the current interpolation point;

[0091] For the first segment of motion trajectory, calculate the jth (i) The pose of the interpolation point is:

[0092]

[0093] in, is the attitude quaternion of the robotic arm at the starting point S, is the quaternion of the robot's posture at the end point T; is the angle between the two quaternions; after the interpolation is completed, Then convert it into a rotation matrix to get the posture

[0094] For the second motion trajectory, due to the requirements of the welding process, the posture of the end of the robot arm usually needs to be constant to ensure the consistency of the welding angle, penetration depth and penetration width. That is, the end posture of the robot arm during the welding process is immutable, and only the position changes with time. Therefore, for the second motion trajectory, the posture of each interpolation point Are limited to R0 is the posture matrix that remains unchanged throughout the welding process and is usually expressed in the form of Euler angles or rotation matrix.

[0095] S23, combining the position and posture of each interpolation point in the motion trajectory into a pose Get the interpolated trajectory sequence of the robotic arm end:

[0096] S3, for each pose of the robot end trajectory sequence Solve the inverse kinematics of the dual manipulator, obtain multiple sets of joint angle solutions and select the optimal solution with coherent motion and minimal deviation Construct a complete joint trajectory sequence

[0097] Specifically, for each pose of the robot end trajectory sequence Use the inverse kinematics algorithm to solve all joint angle combinations that satisfy this pose:

[0098]

[0099] in, represents the jth ( i ) th inverse solution of the nth pose; k is the number of legal inverse solutions existing in the current pose;

[0100] To ensure the smooth continuity of the trajectory in the joint space and away from the singular point, the cost function is set as:

[0101]

[0102] wherein, represents the optimal inverse solution selected by the previous pose; represents the Jacobian matrix corresponding to the joint angle; κ(J) represents the condition number of the Jacobian matrix, which is used to quantify the degree of approaching the singular point; ||·|| represents the Euclidean distance in the joint space, w1 and w2 are weight parameters; σ max (J(θ)) represents the maximum singular value of the Jacobian matrix, σ min (J(θ)) represents the minimum singular value of the Jacobian matrix;

[0103] For the initial pose there is no previous optimal solution, and a set of joint angles θ 1(i) is selected as the initial joint angle;

[0104] For the jth (i) pose , the inverse solution set Θ j(i) is traversed, and the cost function of each inverse solution is calculated , and the optimal solution

[0105]

[0106] The above process is repeated until all poses determine the optimal solution , and a complete joint trajectory sequence is finally constructed

[0107] S4, the joint trajectory sequence of the first segment of the motion trajectory is executed by the dual robot arm, from the initial pose to the welding starting pose

[0108] S5, adjustment is made for the second segment of the motion trajectory:

[0109] S51, the condition number κ(J(θ)) of the Jacobian matrix J(θ) of each point of the joint trajectory sequence is calculated, and potential robot singular poses are analyzed and identified and adjusted.

[0110] Since the system is a six-degree-of-freedom robot, and the posture is limited, it is necessary to avoid entering the singular posture in the entire trajectory. Therefore, the application introduces a singularity judgment mechanism based on the condition number of the Jacobian matrix.

[0111] Specifically, for each joint trajectory point of the joint trajectory sequence By singular value decomposition of the Jacobian matrix J(θ):

[0112] J(θ)=U∑V T

[0113] ∑=diag(σ1,σ2,…,σ n )

[0114] Wherein, U and V are orthogonal matrices; ∑ is a diagonal matrix, and the diagonal elements are singular values σ1, σ2, …, σ n ; When the determinant det(J(θ)) of the Jacobian matrix J(θ) approaches 0 or the smallest singular value is 0 after singular value decomposition of the Jacobian matrix J(θ), it indicates that the robot arm reaches a singular point; In order to judge whether the Jacobian matrix is close to the singular point, its condition number κ(J(θ)) is introduced as the judgment basis of singularity, which is defined as follows:

[0115]

[0116] Wherein, σ max (J(θ)) represents the maximum singular value of the Jacobian matrix, and σ min (J(θ)) represents the minimum singular value of the Jacobian matrix; When the condition number κ(J(θ)) tends to infinity, or σ min (J(θ)) tends to 0, it indicates that the robot arm has approached or entered a singular point, so a singularity judgment threshold κ max is set, if the current condition number κ(J(θ)) exceeds the singularity judgment threshold κ max , it is judged as a potential singular posture, and a speed reduction, avoidance or interpolation adjustment strategy is adopted to ensure the numerical feasibility of the trajectory.

[0117] S52, according to the singularity degree, the required damping factor λ is calculated in real time, and the joint speed of the robot arm is smoothly adjusted to avoid instability when approaching the singular point.

[0118] Specifically, the damping least square method is used to solve the stability of the joint speed:

[0119]

[0120] Wherein, is the joint speed vector; is the velocity vector of the end of the robot arm; λ is a damping factor, I is a unit matrix, the damping factor is used to compensate for the irreversible problem caused by singularity when the minimum singular value is small, and ensure that the matrix JJ T + λ 2 I always remains positive definite and invertible, and the method can effectively suppress the divergence of the solution when approaching the singular point;

[0121] In order to adapt to the numerical characteristics of different postures, the damping factor λ is dynamically adjusted according to the condition number κ(J(θ)):

[0122]

[0123] Where λ0 is the basic damping value, κ safe is the condition number alarm threshold, and δ is the growth index; this strategy makes the system still have good numerical stability and controllability in the singular neighborhood, and ensures that there is no sudden change in speed during the trajectory execution process.

[0124] S53, the trajectory of the double robot arm is analyzed to avoid collision, and the motion speed and timing of the double robot arm are adjusted to avoid interference and ensure synchronous operation when potential collision is detected.

[0125] In the double robot arm welding task, in order to prevent spatial interference between the two robot arms, the invention introduces a double robot arm cooperative collision avoidance mechanism. For the second segment of the motion trajectory, the minimum spatial distance d i , between the two robot arms at each time t 12 (t i ) is calculated;

[0126] When d 12 (t) < d safe at any time, d safe is the set minimum safety distance threshold, and it is judged as potential two-robot arm interference; the motion speed and timing of the two robot arms are fine-tuned to dynamically realize time staggered collision avoidance, that is, without changing the trajectory path and end posture, the two robot arms are staggered to run through rhythm coordination to avoid collision risk and ensure d 12 (t i ) ≥ d safe .

[0127] S6, control the synchronous motion of the double robot arm according to the planned trajectory in the main welding stage, and calculate the joint speed of each joint of the robot arm in real time to ensure that the welding process is smooth and the weld quality meets the requirements.

[0128] S7, after welding, the retreat path of the robot arm is planned and calculated using interpolation method, and the robot arm is controlled to retreat along the planned path smoothly; the double robot arm is reset to the initial position.

[0129] The application provides a trajectory planning and motion control method suitable for a humanoid welding robot, and is especially suitable for realizing high-precision trajectory tracking, singular point avoidance and double-arm collision avoidance control under the constraint condition that the end position and posture are determined and the end posture needs to be kept unchanged during welding.

[0130] Embodiment two

[0131] The embodiment provides a readable storage medium, wherein the readable storage medium stores a computer program, and the computer program causes a processor to execute the welding trajectory planning method of the four-legged humanoid welding robot when the computer program is executed by the processor.

[0132] Embodiment three

[0133] The embodiment provides a computer device, which comprises a processor and a memory for storing a program executable by the processor, and the processor realizes the welding trajectory planning method of the four-legged humanoid welding robot when executing the program stored in the memory.

[0134] The above-mentioned embodiments are the preferred embodiments of the application, but the embodiments of the application are not limited to the above-mentioned embodiments, and any changes, modifications, substitutions, combinations and simplifications made without departing from the spirit and principle of the application should be equivalent replacement modes and should be included in the protection scope of the application.

Claims

1. A welding trajectory planning method for a quadruped humanoid welding robot, characterized by: Based on a quadruped humanoid welding robot with dual robotic arms; the method comprises the following steps: S1, the welding robot navigates to the welding station, identifies and extracts weld information; S2, use interpolation method to plan the motion trajectory of the dual manipulators, and obtain the interpolated manipulator end trajectory sequence; the motion trajectory is divided into two sections, namely, from the initial posture To the welding starting position The first motion trajectory, and the starting position from welding To welding end position The second motion trajectory of i = 1, 2, representing two robotic arms; S3. Solve the inverse kinematics of the dual manipulators for each position in the trajectory sequence of the manipulator end, obtain multiple sets of joint angle solutions and select the optimal solution with consistent motion and minimum deviation. Construct a complete joint trajectory sequence S4, the joint trajectory sequence of the dual robotic arms executing the first motion trajectory; S5. Adjust the second motion trajectory, including: S51. Calculate the condition number κ(J(θ)) of the Jacobian matrix J(θ) at each point in the joint trajectory sequence, and analyze and identify potential singular poses of the manipulator. S52, calculating the required damping factor λ in real time according to the degree of singularity, and smoothly adjusting the joint velocity of the robotic arm when approaching the singularity point; S53, performing collision avoidance analysis on the trajectories of the dual robotic arms, and adjusting the movement speed and timing of the dual robotic arms when a potential collision is detected; S6, the joint trajectory sequence of the dual robotic arms executing the second motion trajectory; S7. After welding is completed, the dual robotic arms are reset to their initial positions.

2. The welding trajectory planning method of a quadruped humanoid welding robot according to claim 1, characterized in that: The step S51 is as follows: For the second segment of motion trajectory, for each joint trajectory point in the joint trajectory sequence By performing singular value decomposition on the Jacobian matrix J(θ): J(θ)=U∑V T ∑=diag(σ1,σ2,…,σ n ) Where U and V are orthogonal matrices; ∑ is a diagonal matrix with singular values ​​σ1, σ2,…, σ n ; Calculate the condition number κ(J(θ)): Among them, σ max (J(θ)) represents the maximum singular value of the Jacobian matrix, σ min (J(θ)) represents the minimum singular value of the Jacobian matrix; set a singularity judgment threshold κ max , if the current condition number κ(J(θ)) exceeds the singularity judgment threshold κ max , it is identified as a potential singular posture, and a deceleration, avoidance or interpolation adjustment strategy is adopted.

3. The welding trajectory planning method for a quadruped humanoid welding robot according to claim 1, characterized in that: The step S52 is to use the damped least square method to solve the stability of the joint velocity for the second motion trajectory: in, is the joint velocity vector; is the velocity vector of the end of the manipulator; λ is the damping factor, and I is the unit matrix; The damping factor λ is dynamically adjusted according to the condition number κ(J(θ)): Among them, λ0 is the basic damping value, κ safe is the condition number warning threshold, and δ is the growth exponent.

4. The welding trajectory planning method for a quadruped humanoid welding robot according to claim 1, characterized in that: The step S53 is to calculate the movement of the two manipulators at each moment t for the second segment of the motion trajectory. i , The minimum spatial distance d 12 (t i ); When there is d at any moment 12 (t) <d safe When d safe The minimum safety distance threshold is set to identify potential interference between the two robotic arms; fine-tune the movement speed and timing of the two robotic arms to ensure d 12 (t i )≥d safe .

5. The welding trajectory planning method of a quadruped humanoid welding robot according to claim 1, characterized in that: The step S2 includes the following steps: S21. Get the initial position of the end of the dual robotic arm and posture The weld information of step S1 includes: weld path and welding posture required for the weld; According to the three-dimensional coordinates of the weld path, the three-dimensional coordinates of the welding start position and the welding end position of the dual robot arm are obtained The posture of the end of the robot arm in the main welding motion trajectory is fixed and restricted to Construct the initial pose of the dual robotic arms Welding starting position and welding termination posture for: S22, respectively, for the initial pose To the welding starting position The first motion trajectory, and the starting position from welding To welding end position The second segment of the motion trajectory is refined using interpolation to calculate the position of each interpolation point. and posture S23, combining the position and posture of each interpolation point in the motion trajectory into a pose Get the interpolated trajectory sequence of the robotic arm end:

6. The welding trajectory planning method for a quadruped humanoid welding robot according to claim 5, characterized in that: The step S22 is to calculate the Euclidean distance L between the starting point S and the end point T of the dual robot end in the motion trajectory. (i) ; Calculate the number of interpolation points required: Among them, d is the set interpolation step size; N (i) The value of should be rounded up; For each interpolation point j (i) =1,2,…,N (i) , calculate the jth (i) The position of the interpolation point for: in, is the position of the starting point S; is the position of the end point T; Indicates the scale of the current interpolation point; For the first segment of motion trajectory, calculate the jth (i) The pose of the interpolation point is: in, is the attitude quaternion of the robot arm at the starting point S, is the quaternion of the robot's posture at the end point T; is the angle between the two quaternions; after the interpolation is completed, Then convert it into a rotation matrix to get the posture For the second segment of the motion trajectory, the posture of each interpolation point Are limited to 7. The welding trajectory planning method for a quadruped humanoid welding robot according to claim 1, characterized in that: The step S3 refers to: For each pose of the robot end trajectory sequence Use the inverse kinematics algorithm to solve all joint angle combinations that satisfy this pose: in, Indicates the jth (i) The nth inverse solution of the pose; k is the number of legal inverse solutions of the current pose; Set the cost function to: in, Indicates the optimal inverse solution selected for the previous pose; represents the Jacobian matrix corresponding to the joint angle; κ(J) represents the condition number of the Jacobian matrix, which is used to quantify the degree of proximity to the singular point; ||·|| represents the Euclidean distance in the joint space, w1 and w2 are weight parameters; σ max (J(θ)) represents the maximum singular value of the Jacobian matrix, σ min (J(θ)) represents the minimum singular value of the Jacobian matrix; For the jth (i) Positions Traverse the inverse solution set Θ j(i) , calculate each inverse solution The cost function Select the optimal solution Repeat the above process until all poses The optimal solution is determined Finally, a complete joint trajectory sequence is constructed 8. A readable storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, enables the processor to execute the welding trajectory planning method for a quadruped humanoid welding robot according to any one of claims 1 to 7.

9. A computer device comprising a processor and a memory for storing a program executable by the processor, characterized in that: When the processor executes the program stored in the memory, the welding trajectory planning method for the quadruped humanoid welding robot according to any one of claims 1 to 7 is implemented.

Images