An adaptive sliding mode control method for the end effector of a wire harness winding robot

Abstract

The present invention discloses an adaptive sliding mode control method for the end effector of a wire harness winding robot. The present invention comprises the following steps: Step 1, establishing a brushless DC motor system model according to Kirchhoff's voltage law, the principle of rigid body dynamics, the principle of backlash dynamics and the law of load variation; Step 2, adaptively identifying system parameters according to the principle of parameter estimation error; Step 3, using the system parameters identified in Step 2 as part of the sliding mode control parameters, and establishing an adaptive fast non-singular terminal sliding mode controller; Step 4, performing online parameter identification and adaptive control on the end effector of the wire harness winding robot based on the initial parameter values established in Step 3. The present invention has the characteristics of strong robustness and fast control response. Ultimately, it can sense load changes in real time and adaptively adjust controller parameters. It is simple and easy to apply, has low cost, and can be widely used in control application fields such as the end effector of a wire harness winding robot.

Description

Technical Field

[0001] The present invention relates to the field of motor control, and in particular to an adaptive sliding mode control method for an end effector of a wire harness winding robot. Background Art

[0002] In the textile industry, cotton yarn and synthetic fiber yarn are often wound into various yarn balls and strands suitable for textile machines. In the electronics industry, enameled copper wire is often used to wind inductors for electrical products, such as various motors, fluorescent lamp ballasts, transformers of various sizes, intermediate frequency coils and inductors for televisions and radios, line output transformers, high-voltage coils for electronic ignition systems and mosquito repellents, voice coils for speakers, headphones, microphones, and various welding machines. In the automotive industry, various wiring harnesses require tape or non-tape wrapping to achieve bundling, leakage protection, aging resistance, and corrosion resistance.

[0003] In the automotive industry, factories need to package various wire harnesses to achieve bundling, leakage protection, aging protection, and corrosion protection. Therefore, to reduce labor intensity, improve efficiency, and modernize production, handheld wire harness winding machines are widely used. Algorithms that can be applied to the end effector of a wire harness winding robot include PID, neural network control, and robust control. The PID method is simple and easy to apply, but is insensitive to environmental changes. Neural network control methods do not require specific models, but require a large amount of calculation and are not suitable for controllers with weaker performance. Robust control prioritizes stability and reliability and can operate within a certain range, but may not be optimal, resulting in less than ideal control accuracy.

[0004] At present, in order to overcome the problem that the controller of the end effector of the wire harness winding robot cannot perceive external interference caused by load changes, the present invention discloses a fast non-singular terminal sliding mode control method based on parameter estimation for the end effector of the wire harness winding robot, which has wide application value. Summary of the Invention

[0005] The object of the present invention is to provide an adaptive sliding mode control method for an end effector of a wire harness winding robot in view of the deficiencies in the prior art.

[0006] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0007] Step 1: Establish a brushless DC motor system model based on Kirchhoff's voltage law, rigid body dynamics principle, backlash dynamics principle and load change law.

[0008] Step 2: Perform adaptive identification of system parameters based on the parameter estimation error principle.

[0009] Step 3: Use the system parameters identified in step 2 as part of the sliding mode control parameters, and establish an adaptive fast non-singular terminal sliding mode controller.

[0010] Step 4: Based on the initial parameter values established in step 3, online parameter identification and adaptive control are performed on the end effector of the harness winding robot;

[0011] Furthermore, the specific implementation of step 1 includes the following sub-steps:

[0012] (1.1) Establish the brushless DC motor system model, the formula is as follows:

[0013]

[0014] Where u is the input voltage, R is the equivalent resistance, J m is the motor moment of inertia, J l is the load moment of inertia, ω l is the load speed, Indicates load speed ω l The derivative of T l is the load torque, d is the disturbance and unmodeled dynamic part, K e is the electromotive force constant, K t is the motor torque constant, B m is the motor damping constant, B l is the load damping constant, and n is the transmission ratio.

[0015] (1.2) Establish a load variation function based on step (1.1);

[0016] T l =-K l θ l (2)

[0017] Among them, K l >0 is the load variation coefficient, θ l is the load angular displacement.

[0018] (1.3) Based on steps (1.1) and (1.2), a linear regression motor system model is established as follows:

[0019]

[0020] To facilitate subsequent derivation and calculation, the parameter x2 is used to replace the load speed ω l ; where Θ is the coefficient vector, θ1, θ2, θ3, θ4 are system parameters, ψ is the regression vector, and B is the damping coefficient, as shown below:

[0021]

[0022] ψ=[-x2,u,x1,-1] T (5)

[0023] To facilitate subsequent derivation and calculation, the parameter x1 in formula (5) represents the load angular displacement θ l ;

[0024] The step (2) includes the following sub-steps:

[0025] (2.1) Design a system parameter identification method based on parameter estimation error, constructed based on step (1.3).

[0026] First, perform first-order filtering, the formula is as follows:

[0027]

[0028] Among them, k is the filter coefficient, x 2f and ψ f is the filtered value, Respectively represent x 2f and Ψ f Take the derivative; x 2f It can be expressed by the following formula:

[0029]

[0030] Secondly, design an auxiliary matrix M, auxiliary vector N and auxiliary vector H, which are expressed as follows:

[0031]

[0032] Where, l is the filter coefficient; They represent the derivatives of the auxiliary matrix M and the auxiliary vector N respectively.

[0033] Then, construct the auxiliary vector H:

[0034]

[0035] in, is the estimated value of the system parameter, so the auxiliary vector H includes the parameter estimation error information.

[0036] Finally, the gain adaptive update law is designed

[0037]

[0038] Where Γ represents the error estimation gain value, and ρ represents the gain coefficient.

[0039] (2.2) Based on step (2.1), the system parameter estimation differential expression based on the parameter estimation error information is obtained As shown below:

[0040]

[0041] Furthermore, the step (3) establishes an adaptive fast non-singular terminal sliding mode controller, which is specifically implemented as follows:

[0042] Take the sliding surface function s as follows:

[0043]

[0044] Take the reaching law function As described below:

[0045]

[0046] Combining the above two equations, we get the following:

[0047]

[0048] Where, u is the system input voltage, is the estimated value of the system parameters, coefficients k1>0, k2>0, α>1, k3>0, k4>0, 1>β>0, is the derivative value of the set speed.

[0049] Furthermore, the step (4) is specifically as follows: first, a system model of the controlled object is obtained through step (1); secondly, the unknown coefficients of the system in step (1) are identified online using step (2); and finally, the rotation speed of the end effector of the harness winding robot is controlled by the controller designed in step (3). Finally, this method can achieve the system parameters obtained by online identification based on the input and output values and adjust some parameters of the controller in real time. Compared with other methods, it has the characteristics of fast response speed and high control accuracy.

[0050] The beneficial effects of the present invention are as follows:

[0051] In terms of speed control applications for the end effector of a wire harness winding robot, compared with existing methods, the method of the present invention is based on the parameter estimation error information proposed in step (2) to estimate system parameters, has a faster identification speed, and can quickly and adaptively adjust the controller. Based on the fast non-singular terminal sliding mode control method in step (3), it has the characteristics of strong robustness and fast control response. Ultimately, it achieves real-time perception of load changes and adaptive adjustment of controller parameters. It is simple and easy to apply, low-cost, and can be widely used in control application fields such as the end effector of a wire harness winding robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is the control algorithm block diagram

[0053] Figure 2 This is the parameter identification diagram of the motor system of the end effector of the harness winding robot

[0054] Figure 3 This is the speed tracking diagram of the harness winding robot end effector motor system under the control of the adaptive fast non-singular terminal sliding mode controller. DETAILED DESCRIPTION

[0055] The present invention will be further described below with reference to the accompanying drawings.

[0056] The purpose of the present invention is to design an adaptive fast non-singular terminal sliding mode controller using parameter estimation error information, which can adaptively control the end effector of a wire harness winding robot and has good working effect.

[0057] like Figure 1 The control algorithm block diagram of the embodiment of the present invention is shown in FIG. Algorithm step (1) includes the following sub-steps:

[0058] (1.1) Establish the brushless DC motor system model, the formula is as follows:

[0059]

[0060] Where u is the input voltage, R is the equivalent resistance, J m is the motor moment of inertia, J l is the load moment of inertia, ω l is the load speed, Indicates load speed ω l The derivative of T l is the load torque, d is the disturbance and unmodeled dynamic part, K e is the electromotive force constant, K t is the motor torque constant, B m is the motor damping constant, B l is the load damping constant, and n is the transmission ratio.

[0061] (1.2) Establish a load variation function based on step (1.1);

[0062] T l =-K l θ l (16)

[0063] Among them, K l >0 is the load variation coefficient, θ l is the load angular displacement.

[0064] (1.3) Based on steps (1.1) and (1.2), a linear regression motor system model is established as follows:

[0065]

[0066] To facilitate subsequent derivation and calculation, the parameter x2 is used to replace the load speed ω l ; where Θ is the coefficient vector, Ψ is the regression vector, and B is the damping coefficient, as shown below:

[0067]

[0068] Ψ=[-x2,u,x1,-1] T (19)

[0069] To facilitate subsequent derivation and calculation, the parameter x1 is used to replace the load angular displacement θ l ;

[0070] like Figure 2 As shown in FIG, the parameter identification process of the motor system of the end effector of the online harness winding robot based on the parameter estimation error is shown. The specific step (2) includes the following sub-steps:

[0071] (2.1) Design a system parameter identification method based on parameter estimation error, constructed based on step (1.3).

[0072] First, perform first-order filtering, the formula is as follows:

[0073]

[0074] Among them, k is the filter coefficient, x 2f and Ψ f is the filtered value, Indicates x 2f and Ψ f Take the derivative; x 2f It can be expressed by the following formula:

[0075]

[0076] Secondly, design an auxiliary matrix M, auxiliary vector N and auxiliary vector H, which are expressed as follows:

[0077]

[0078] Where l is the filter coefficient;

[0079] Then, construct the auxiliary vector H:

[0080]

[0081] in, is the estimated value of the system parameter, so the auxiliary vector H includes the parameter estimation error information.

[0082] Finally, the gain adaptive update law is designed:

[0083]

[0084] Where Γ represents the error estimation gain value, and ρ represents the gain coefficient.

[0085] (2.2) Based on the basic step (2.1), the differential expression of the system parameter estimation based on the parameter estimation error information is obtained As shown below:

[0086]

[0087] Furthermore, the step (3) establishes an adaptive fast non-singular terminal sliding mode controller.

[0088] Take the sliding surface function s as follows

[0089]

[0090] Take the reaching law function As described below

[0091]

[0092] Combining the above two equations, we get the following:

[0093]

[0094] Where, u is the system input voltage, is the estimated value of the system parameters, coefficients k1>0, k2>0, α>1, k3>0, k4>0, 1>β>0, is the derivative value of the set speed.

[0095] Furthermore, the step (4) is specifically as follows: first, the controlled object system model is obtained through step (1); secondly, the unknown coefficients of the system in step (1) are identified online using step (2); and thirdly, the speed of the end effector of the harness winding robot is controlled by the controller designed in step (3). Finally, this method can achieve the system parameters obtained by online identification based on the input and output values and adjust some controller parameters in real time. Compared with other methods, it has the characteristics of fast response speed and high control accuracy. Figure 3 Figure 2 shows the speed tracking process of the end-effector motor system of the harness winding robot under the control of an adaptive fast non-singular terminal sliding mode controller based on parameter identification.

[0096] Example 1:

[0097] During the operation of a wire harness winding robot's end effector, the process requirements for point winding require the actuator to respond quickly and achieve ideal winding results. Currently, the PID method can achieve a certain speed control effect under normal circumstances. However, during the point winding process, where frequent acceleration and deceleration occur and the load gradually decreases, overshoot and jitter of approximately 10% occur, resulting in unstable speed control. Furthermore, additional control parameter adjustments are required for different product batches. Compared to the existing methods, this method can quickly identify system parameter information during the control process and adaptively adjust controller parameters, resulting in only approximately 1% overshoot during acceleration and deceleration. When a step speed signal is set to 225 rpm, the speed reaches 242 rpm and the voltage reaches 12.54 V under PID control. However, the proposed method achieves a peak speed of 228 rpm and a voltage of 11.35 V. Therefore, this method achieves fast response and minimal overshoot. It has greater applicability and effectively achieves speed control results.

Claims

1. An adaptive sliding mode control method for the end effector of a harness winding robot, characterized in that The steps include: Step 1: Establish a brushless DC motor system model based on Kirchhoff's voltage law, rigid body dynamics, backlash dynamics, and load variation law; Step 2: Adaptively identify system parameters based on the parameter estimation error principle; Step 3: Use the system parameters identified in step 2 as part of the sliding mode control parameters, and establish an adaptive fast non-singular terminal sliding mode controller; Step 4: Based on the initial parameter values established in step 3, online parameter identification and adaptive control are performed on the end effector of the harness winding robot; The specific implementation of step 1 includes the following sub-steps: (1.1) Establish the brushless DC motor system model, the formula is as follows: Where u is the input voltage, R is the equivalent resistance, J m is the motor moment of inertia, J l is the load moment of inertia, ω l is the load speed, Indicates load speed ω l The derivative of T l is the load torque, d is the disturbance and unmodeled dynamic part, K e is the electromotive force constant, K t is the motor torque constant, B m is the motor damping constant, B l is the load damping constant, n is the transmission ratio; (1.2) Establish a load variation function based on step (1.1); T l =-K l i l (2) Among them, K l >0 is the load variation coefficient, θ l is the load angular displacement; (1.3) Based on steps (1.1) and (1.2), a linear regression motor system model is established as follows: To facilitate subsequent derivation and calculation, the parameter x2 is used to replace the load speed ω l ; where Θ is the coefficient vector, θ1, θ2, θ3, θ4 are system parameters, ψ is the regression vector, and B is the damping coefficient, as shown below: Ψ=[-x2,u,x1,-1] T (5) To facilitate subsequent derivation and calculation, the parameter x1 in formula (5) represents the load angular displacement θ l ; The step 2 includes the following sub-steps: (2.1) Design a system parameter identification method based on parameter estimation error, constructed based on step (1.3); First, perform first-order filtering, the formula is as follows: Among them, k is the filter coefficient, x 2f and ψ f is the filtered value, Respectively represent x 2f and Ψ f Take the derivative; x 2f It can be expressed by the following formula: Secondly, design an auxiliary matrix M, auxiliary vector N and auxiliary vector H, which are expressed as follows: Where, l is the filter coefficient; Respectively represent the derivatives of the auxiliary matrix M and the auxiliary vector N; Then, construct the auxiliary vector H: in, is the estimated value of the system parameter, so the auxiliary vector H includes the parameter estimation error information; Finally, the gain adaptive update law is designed Where Γ represents the error estimation gain value, and ρ represents the gain coefficient; (2.2) Based on step (2.1), the system parameter estimation differential expression based on the parameter estimation error information is obtained As shown below: Step 3 establishes an adaptive fast non-singular terminal sliding mode controller, which is specifically implemented as follows: Take the sliding surface function s as follows: Take the reaching law function As described below: Combining the above two equations, we get the following: Where, u is the system input voltage, is the estimated value of the system parameters, coefficients k1>0, k2>0, α>1, k3>0, k4>0, 1>β>0, is the derivative value of the set speed; The step 4 is specifically implemented as follows: First, the system model of the controlled object is obtained through step 1. Then, the unknown coefficients of the system in step 1 are identified online through step 2. Then, the speed of the end effector of the harness winding robot is controlled by the controller designed in step 3. Finally, the system parameters can be identified online based on the input and output values, and some controller parameters can be adjusted in real time.

Images