A dual-joint robot arm vibration suppression method and system based on torque decoupling

Abstract

The application discloses a double-joint robot vibration suppression method and system based on torque decoupling, which comprises the following steps: constructing a motor end dynamics nominal model, and solving a theoretical output torque of a joint input torque on a single joint in real time; measuring an actual joint torque affected by coupling force, calculating a difference value between the theoretical output torque and the actual joint torque, and the difference value is an estimated coupling force between joints; filtering the estimated coupling force by using a low-pass filter to obtain a coupling component between the joints, introducing a joint control loop through a negative feedback mechanism, forming feedforward compensation, superimposing the coupling component and the original input torque, dynamically canceling the coupling interference torque, realizing torque decoupling, and suppressing vibration. The method does not depend on the accurate analysis of the dynamics model of the double-joint robot body, is less sensitive to the precision of model parameters, has a simple calculation process, has strong engineering applicability, and is beneficial to the implementation of the robot vibration suppression strategy.

Description

Technical Field

[0001] This invention relates to the field of vibration suppression for robotic arms, and in particular to a method and system for vibration suppression of a dual-joint robotic arm based on torque decoupling. Background Technology

[0002] One of the development trends of industrial robotic arms is lightweighting and flexibility. However, due to the presence of flexible joints in the robotic arm, residual vibrations can occur due to abrupt changes in motion after the end effector reaches the desired position, affecting the position control accuracy and performance of the robotic arm. In engineering, motion trajectory optimization strategies are most commonly used to suppress residual vibrations, with typical strategies including input shaping and S-curve optimization. Considering the nonlinear and strongly coupled characteristics of the dynamic model of a flexible joint robotic arm, the vibration suppression effect of motion trajectory optimization strategies is affected by the coupling torque components. Therefore, decoupling operations between joints are beneficial to improving vibration suppression.

[0003] For dual-joint robotic arms, the traditional feedforward decoupling method based on system dynamics models is an open-loop control strategy. Its decoupling matrix needs to be solved based on the system's transfer function, which involves significant computational complexity and is highly dependent on the accuracy of model parameters. When parameter accuracy is low, the decoupling effect is poor, thus affecting the robotic arm's vibration suppression performance. Therefore, it is necessary to develop a decoupling method with a simpler computational process and lower dependence on model parameter accuracy. Summary of the Invention

[0004] In order to overcome the above-mentioned shortcomings and deficiencies of the prior art, the purpose of this invention is to provide a vibration suppression method and system for a dual-joint robotic arm based on torque decoupling, which utilizes joint torque sensors and coupled torque observers to achieve torque decoupling, thereby solving the technical problems mentioned in the background art.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] A vibration suppression method for a dual-joint robotic arm based on torque decoupling includes:

[0007] Construct a nominal dynamic model of the motor end and calculate in real time the theoretical output torque generated by the joint input torque on a single joint;

[0008] The actual joint torque affected by the coupling force is measured, and the difference between the theoretical output torque and the actual joint torque is calculated. The difference is the estimated coupling force between the joints.

[0009] A low-pass filter is used to filter the estimated coupling force to obtain the coupling component between joints. A negative feedback mechanism is introduced into the joint control loop to form feedforward compensation, so that the coupling component and the original input torque are vector-superimposed to achieve dynamic cancellation of the coupling interference torque, realize torque decoupling, and suppress vibration.

[0010] Furthermore, a nominal dynamic model for the motor end is constructed, specifically as follows:

[0011]

[0012] in:

[0013] G τi (s), i = 1, 2 represents the input torque τ of the i-th joint. mi (s) to the theoretical output torque The transfer function;

[0014]

[0015] J mi ,b mi ,k si ,b si ,b li ,i=1,2 represent the moment of inertia, motor end damping coefficient, stiffness coefficient, transmission damping coefficient, and load end damping coefficient of the i-th joint, respectively; J l11 J l22 These represent the moments of inertia at the load ends of the first and second joints, respectively:

[0016]

[0017] M li ,L li , i = 1, 2 represent the mass and length of the i-th joint, respectively, θ l2 It is the included angle between the two links of the dual-joint robotic arm.

[0018] Furthermore, the torque sensors are respectively installed at two joints of the robotic arm.

[0019] Furthermore, the original input torque is obtained by controlling the joint setting angle through a control loop.

[0020] Furthermore, the control loop includes a position loop controller and a velocity loop controller, and the control loop employs an AS curve motion trajectory optimization strategy to suppress residual vibration.

[0021] A system for implementing the vibration suppression method of the dual-joint robotic arm as claimed in the claims includes: a control loop and a dual-joint coupling torque observer, wherein the dual-joint coupling torque observer includes a torque decoupling loop;

[0022] The torque decoupling circuit is used to construct the nominal dynamic model of the motor end and to calculate the theoretical output torque generated by the joint input torque on a single joint in real time.

[0023] The actual joint torque affected by the coupling force is measured, and the difference between the theoretical output torque and the actual joint torque is the estimated coupling force between the joints.

[0024] A low-pass filter is used to filter the estimated coupling force to obtain the coupling component between joints, and a negative feedback mechanism is introduced into the joint control loop.

[0025] A storage medium storing a computer program that, when executed by a processor, implements the vibration suppression method for the dual-joint robotic arm.

[0026] A device includes a memory, a processor, and the vibration suppression method for a dual-joint robotic arm stored in the memory and operable on the processor.

[0027] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0028] 1. This invention decouples the dual-joint robotic arm from a dual-input dual-output system into two single-input single-output systems. Then, it uses AS curves to suppress residual vibration in the two decoupled control loops. Compared with the undecoupled system, this invention significantly improves the vibration suppression capability.

[0029] 2. This invention uses the nominal dynamic model of the joint motor end in a dual-joint robotic arm to estimate the theoretical output torque. This model is then used to perform vector subtraction with the composite torque measured by the joint torque sensor to obtain the coupled torque component. This coupled torque component is then introduced into the motion control loop through negative feedback to achieve dynamic cancellation of the coupled torque. The implementation process does not rely on a precise dual-joint robotic arm dynamic model, has low computational complexity, and is simple and reliable.

[0030] 3. This invention acquires real-time output data from the torque sensor, directly observing the dynamic coupling effect through physical quantities. The decoupled torque components are then vector-superimposed with the original input torque of the motion control loop—a negative feedback closed-loop control—to improve the vibration suppression effect of the robotic arm. The model parameters have low precision dependence, making it easier to improve the decoupling effect and effectively suppress residual vibration degradation caused by model mismatch. Attached Figure Description

[0031] Figure 1 This is a system block diagram of a vibration suppression method for a dual-joint robotic arm based on torque decoupling provided in an embodiment of the present invention;

[0032] Figure 2 A schematic diagram illustrating the principle of a vibration suppression method for a dual-joint robotic arm based on torque decoupling provided in an embodiment of the present invention.

[0033] Figures 3(a), 3(b), 3(c), and 3(d) are comparison diagrams of joint angles and end-effector vibration acceleration waveforms during long-stroke unloaded operation of the embodiments of the present invention under the conditions of no decoupling, using torque decoupling method, and using system dynamics model decoupling method. Figure 3(a) shows the mechanical angle of the first joint, and Figure 3(b) shows the mechanical angle of the second joint. Both mechanical angles are rotation angles relative to zero mechanical degrees. Figure 3(c) shows the vibration acceleration of the end-effector of the robotic arm in the X direction, and Figure 3(d) shows the vibration acceleration of the end-effector of the robotic arm in the Y direction.

[0034] Figures 4(a), 4(b), 4(c), and 4(d) are comparison diagrams of joint angles and end-effector vibration acceleration waveforms during long-stroke operation under load in embodiments of the present invention, with no decoupling, torque decoupling method, and system dynamics model decoupling method, respectively. Figure 4(a) shows the mechanical angle of the first joint, and Figure 4(b) shows the mechanical angle of the second joint. Both mechanical angles are rotation angles relative to zero mechanical degrees. Figure 4(c) shows the X-direction vibration acceleration of the end-effector of the robotic arm, and Figure 4(d) shows the Y-direction vibration acceleration of the end-effector of the robotic arm. Detailed Implementation

[0035] The present invention will be further described in detail below with reference to the embodiments, but the implementation of the present invention is not limited thereto.

[0036] Example

[0037] like Figures 1-2 As shown, this invention obtains the coupling relationship between joints in the dynamic model of a dual-joint robotic arm, decouples the coupling components between joints, and introduces them into the control loop to achieve dynamic cancellation of the coupling components, thereby reducing the impact of the coupling relationship between joints on vibration suppression and thus better implementing the vibration suppression strategy.

[0038] This embodiment discloses a vibration suppression method for a dual-joint robotic arm based on torque decoupling, including:

[0039] Using the Lagrange equations and neglecting the Coriolis force and centrifugal force, we can obtain the dynamic model equations for the two-joint robotic arm:

[0040]

[0041] Where, τ m =[τ m1 τ m2 ] T θ m =[θ m1 θ m2 ] T θ l =[θ l1θ l2 ] T These are the input torque, the angle vectors at the motor end, and the load end, respectively; J m =diag(J m1 J m2 ), It is a 2×2 inertia matrix; B S =diag(b S1 ,b S2 ), B m =diag(b m1 ,b m2 ), B l =diag(b l2 ,b l2 K is a 2×2 damping matrix. S =diag(k) S1 ,k S2 τ is a 2×2 elasticity coefficient matrix. mi ,θ mi ,θ li J represents the input torque, motor rotation angle, and load rotation angle of the i-th joint, respectively; mi ,b mi ,b si ,b li ,k si ,i=1,2 represent the moment of inertia of the i-th joint, the damping coefficient at the motor end, the damping coefficient at the transmission end, the damping coefficient at the load end, and the stiffness coefficient, respectively; J l The elements of the inertia matrix are respectively

[0042]

[0043] M li ,L li , i = 1, 2 represent the mass and length of the i-th joint, respectively, θ l2 It is the included angle between the two links of the dual-joint robotic arm.

[0044] The dynamic model equations of the dual-joint robotic arm are rearranged as follows:

[0045]

[0046] On the right-hand side of the first equation, the acceleration equation at the load end of the first joint consists of two parts: the first part... The uncoupled part has parameters that depend only on the first joint; the second part... The coupling part indicates that during motion, a portion of the acceleration of the second joint is coupled to the acceleration of the first joint. Similarly, on the right side of the second equation, it can be deduced that during motion, a portion of the acceleration of the first joint is coupled to the acceleration of the second joint.

[0047] The results of rearranging the terms show that there is a coupling effect in the torque.

[0048] like Figure 1 As shown, this invention obtains and feeds back the total joint torque through joint sensors in the robotic arm. Simultaneously, it calculates in real-time the theoretical output torque generated by the input torque on a single joint. Then, it performs a vector subtraction operation between the theoretical output torque and the joint feedback torque to separate the dynamic coupling torque components between the joints. A low-pass filter is used to perform phase compensation and noise suppression on the estimated coupling force. The filtered and optimized coupling force estimate is introduced into the joint control loop through a negative feedback mechanism, forming a feedforward compensation structure. This compensated torque is vector-superimposed with the original input torque at the drive end, achieving dynamic cancellation of coupling interference torque through torque closed-loop control. After decoupling, the motion control of each joint can be performed separately using velocity loops and position loops. Based on the set angle of the joint, the AS curve is used for trajectory planning to reduce the vibration of the robotic arm. Decoupling is for better vibration suppression.

[0049] Specifically, the steps include the following:

[0050] like Figure 2 As shown, in order to achieve decoupled control of the system, a coupled torque observer is used to estimate the coupling force between joints. The coupled torque observer is a virtual structure that represents the torque decoupling process.

[0051] A nominal dynamic model of the motor end is constructed to calculate the theoretical output torque generated by the input torque in a single joint body in real time. m1 (s), G m2 (s), G l1 (s), G l2 (s) represents the physical model of the actual robotic arm, with coupling torques present in between; G τ1 (s) represents the input torque τ of the first joint. m1 To its theoretical joint torque The nominal model, G τ2 (s) represents the input torque τ of the second joint. m2 To its theoretical joint torque The nominal model.

[0052] According to the transmission model of the joint, G τ1 (s) and G τ2 (s) can be expressed by the dynamic model as follows:

[0053]

[0054] in

[0055]

[0056] J mi ,b mi ,k si ,b si ,b li ,i=1,2 represent the moment of inertia, motor end damping coefficient, stiffness coefficient, transmission damping coefficient, and load end damping coefficient of the i-th joint, respectively; J l11 J l22 These represent the moments of inertia at the load ends of the first and second joints, respectively:

[0057]

[0058] M li ,L li , i = 1, 2 represent the mass and length of the i-th joint, respectively, θ l2 It is the included angle between the two links of the dual-joint robotic arm.

[0059] The actual joint torque affected by the coupling force is measured, and the difference between the theoretical output torque and the actual joint torque is the estimated coupling force between the joints.

[0060] Specifically:

[0061] The actual total joint torque τ after the influence of coupling force is obtained by measuring the joint torque sensor. l1 and τ l2 Coupling forces between joints and Through the total joint torque τ l1 and τ l2 With torque observation model G τ1 (s) and G τ2 (s) Estimated theoretical joint torque and The difference is:

[0062]

[0063] like Figure 2 As shown, a low-pass filter G was used to improve system robustness. filter1 (s) and G filter2(s) Phase compensation and noise suppression are performed on the estimated coupling force. The filtered and optimized coupling force estimate is introduced into the joint control loop through a negative feedback mechanism to form a feedforward compensation structure. This compensation torque is vector-superimposed with the original input torque at the drive end, and dynamic cancellation of coupling interference torque is achieved through torque closed-loop control.

[0064] After torque decoupling of the dual-joint robotic arm, the position loop controller C... pos (s) and speed loop controller C spd (s) Motion control of the two joints is performed. Furthermore, residual vibration is suppressed by using an asymmetric S (AS) curve motion trajectory optimization strategy, thereby improving the position control accuracy and working performance of the dual-joint robotic arm.

[0065] This embodiment also provides a storage medium on which a computer program is stored, which, when executed by a processor, implements the vibration suppression method for the dual-joint robotic arm.

[0066] This embodiment also provides a device, including a memory, a processor, and the vibration suppression method for the dual-joint robotic arm stored in the memory and capable of running on the processor.

[0067] like Figures 3(a)-3(d) He Ru Figures 4(a)-4(d) As shown in the figure, using the torque decoupling-based vibration suppression method for a dual-joint robotic arm provided in this embodiment, under the same conditions except for the different decoupling methods, the residual vibration suppression effect of the dual joints is significantly better than that without decoupling when the arm travels to a given position with both the end effector unloaded and loaded. Furthermore, the dynamics-based decoupling method relies on accurate model parameters. Deviations between the model parameters and actual values ​​weaken the decoupling effect and thus affect the residual vibration suppression effect. Therefore, the comparative experiments also show that the torque decoupling method is less sensitive to model parameters and more robust.

[0068] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the embodiments described above. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A vibration suppression method for a dual-joint robotic arm based on torque decoupling, characterized in that, include: Construct a nominal dynamic model of the motor end and calculate in real time the theoretical output torque generated by the joint input torque on a single joint; The actual joint torque affected by the coupling force is measured, and the difference between the theoretical output torque and the actual joint torque is calculated. The difference is the estimated coupling force between the joints. A low-pass filter is used to filter the estimated coupling force to obtain the coupling component between joints. A negative feedback mechanism is introduced into the joint control loop to form feedforward compensation, so that the coupling component and the original input torque are vector-superimposed to achieve dynamic cancellation of the coupling interference torque, realize torque decoupling, and suppress vibration. Construct the nominal dynamic model of the motor end, specifically as follows: in: Indicates the first Input torque of each joint To the theoretical output torque The transfer function; They represent the first The moment of inertia of each joint, the damping coefficient at the motor end, the stiffness coefficient, the damping coefficient of the transmission part, and the damping coefficient at the load end; These represent the moments of inertia at the load ends of the first and second joints, respectively: They represent the first The mass and length of each joint It is the included angle between the two links of the dual-joint robotic arm.

2. The vibration suppression method for a dual-joint robotic arm according to claim 1, characterized in that, Torque sensors are installed at two joints of the robotic arm.

3. The vibration suppression method for a dual-joint robotic arm according to claim 1, characterized in that, The original input torque is calculated by the control loop based on the joint setting angle.

4. The vibration suppression method for a dual-joint robotic arm according to claim 3, characterized in that, The control loop includes a position loop controller and a velocity loop controller, and the control loop adopts an AS curve motion trajectory optimization strategy to suppress residual vibration.

5. A system for implementing the vibration suppression method for a dual-joint robotic arm according to any one of claims 1-4, characterized in that, include: A control loop and a dual-joint coupled torque observer, wherein the dual-joint coupled torque observer includes a torque decoupling loop; The torque decoupling circuit is used to construct the nominal dynamic model of the motor end and to calculate the theoretical output torque generated by the joint input torque on a single joint in real time. The actual joint torque affected by the coupling force is measured, and the difference between the theoretical output torque and the actual joint torque is calculated. The difference is the estimated coupling force between the joints. A low-pass filter is used to filter the estimated coupling force to obtain the coupling component between joints, and a negative feedback mechanism is introduced into the control loop.

6. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the vibration suppression method for a dual-joint robotic arm as described in any one of claims 1-4.

7. A device, characterized in that, It includes a memory, a processor, and a vibration suppression method for a dual-joint robotic arm as described in any one of claims 1-4, which is stored in the memory and can run on the processor.

Images