Bionic robot flexible joint angular displacement tracking control method based on terminal sliding mode

Through the combination of Bouc-Wen model and fuzzy state observer, an adaptive terminal sliding mode controller is designed, which solves the problems of hysteresis nonlinearity and unpredictable state in bionic robots, and achieves fast and accurate angles in the presence of unpredictable and disturbances Displacement tracking control improves the robustness and stability of the system.

CN118915428APending Publication Date: 2024-11-08CHINA NORTH VEHICLE RES INST
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410800374.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-20
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The existing intelligent material drive systems have hysteresis nonlinear problems in bionic robots, resulting in insufficient tracking accuracy and stability. When the system state is unpredictable and external disturbances exist, the control performance is difficult to meet the requirements.

Method used

The Bouc-Wen model is used to characterize the hysteresis characteristics, design the fuzzy state observer and adaptive terminal sliding mode controller, and combine the sliding mode control theory and inversion technology to realize the fast and accurate tracking control of the flexible joint angular displacement of the bionic robot, reducing the controller Design difficulty and calculation burden.

Benefits of technology

In the absence of unpredictable system status and external disturbance, the fast and accurate angular displacement tracking control of the flexible joint rotation driver of the bionic robot is realized, which improves the system's robustness and tracking performance and ensures the stability of the closed-loop system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118915428A_ABST
    Figure CN118915428A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of robot joint driving control, in particular to a bionic robot flexible joint angular displacement tracking control method based on a terminal sliding mode. The method comprises the steps of establishment of a flexible joint rotation driver hysteresis nonlinear system model and design of a state observer and a terminal sliding mode controller. The research object is bionic robot flexible joint angular displacement tracking control based on intelligent materials, two performances of observation errors and tracking errors are selected, then a state observer and a self-adaptive terminal sliding mode controller are designed, it is ensured that all signals in a closed-loop system are bounded, and the tracking precision is improved. The fast and accurate angular displacement tracking control of the hysteresis nonlinear system of the flexible joint rotation driver of the bionic robot and the strong robustness of the system under the specific working condition are realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of robot joint drive control, and in particular to a bionic robot flexible joint angular displacement tracking control method based on terminal sliding mode. Background Art

[0002] Smart materials are a new type of functional materials that integrate perception, drive and control. This type of material can sense changes in the external environment and make corresponding changes and drives, and can realize the "in-situ drive without mechanical joints" mode. It has great application value in the fields of bionic robots, national defense, industrial production, biomedicine, etc. Among them, bionic robots are an important research achievement that applies bionics to the field of robotics. Smart materials, as ideal materials in the field of bionic robot drive technology, have been widely used and studied, mainly including piezoelectric ceramics, electroactive polymers, shape memory alloys, and giant magnetostrictive materials. Although the above smart materials have been widely used and studied in the field of bionic robot drive, their inherent hysteresis nonlinearity seriously affects the tracking accuracy and stability of the system. Therefore, eliminating the influence of hysteresis nonlinearity on the smart material drive system is an urgent problem to be solved.

[0003] According to the different modeling methods, hysteresis models can be roughly divided into two categories: physical models and phenomenological models. Among them, the physical model is a hysteresis nonlinear model based on the material's own mechanism of action, such as the Jiles-Atherton model, but its model fitting accuracy is low and its applicability is poor. The phenomenological model describes the hysteresis characteristics based on the relationship between the input and output of the drive system. Commonly used models include the Preisach model, the PI model, the Bouc-Wen model, and the Backlash-like model. The present invention adopts the Bouc-Wen model. Compared with other phenomenological models, the model has relatively few parameters and can separate the hysteresis term.

[0004] At present, the control technology of hysteresis nonlinear system based on smart materials can be roughly divided into open-loop control and closed-loop control. Open-loop control can also be called open-loop feedforward inverse compensation control. This type of control method is highly dependent on the accuracy of the system model. However, in actual engineering applications, accurate system models are not easy to obtain. Therefore, under the action of this type of control method, the hysteresis system cannot obtain the expected control performance. In order to avoid the above problems, closed-loop control was proposed. This type of control method can also be called feedback control method. According to whether hysteresis inverse compensation is required, it can be further divided into feedback control method based on inverse compensation and feedback control method without inverse compensation. The closed-loop control method not only eliminates the hysteresis characteristics of the system, but also improves the tracking performance of the smart material drive system. However, the existing control technology of hysteresis nonlinear system based on smart materials still has shortcomings: most control methods are designed to assume that the upper bound information of the system disturbance is known, and also require that the system state is completely measurable. Summary of the invention

[0005] In view of this, the present invention proposes a bionic robot flexible joint angular displacement tracking control method based on terminal sliding mode, which can realize fast and accurate robust angular displacement tracking control of the bionic robot flexible joint rotation drive system under factors such as unpredictable system state and external disturbances.

[0006] To achieve the above object, the technical solution of the present invention is:

[0007] A bionic robot flexible joint angular displacement tracking control method based on terminal sliding mode, the tracking control method comprising the following steps:

[0008] S1. Based on the BoucWen model to characterize the hysteresis characteristics of the system, a hysteresis nonlinear system model of the bionic robot flexible joint actuator is established;

[0009] S2. Approximate the unknown nonlinear function of the system through the fuzzy logic system, design the fuzzy state observer, and estimate the amount of information of the unmeasurable system state;

[0010] S3. Based on the fuzzy state observer, combined with the inversion technology, adaptive technology and sliding mode control theory, an adaptive terminal sliding mode controller is designed;

[0011] S4. Use Lyapunov stability theory to analyze the rationality of the designed adaptive terminal sliding mode controller and the stability of the closed-loop system.

[0012] The mathematical model of the revolute joint drive system in step S1 is:

[0013]

[0014] Where x0 and x are the initial length and the length after deformation of the displacement actuator, respectively; L0 and L are the initial length and the length after deformation of the spring, respectively; k1 is the spring constant; m, k, c represent mass, damping coefficient, and elastic coefficient respectively; I is the moment of inertia of AB about the midpoint O; w is the stress generated by voltage; is the fluid resistance moment;

[0015] Let x1 = θ, a2=C / I,a3=-ξ / I,a4=-l1 / I,the mathematical model is rewritten as:

[0016]

[0017] Where d1 is the external disturbance of the system, w(u(t)) is the output of the unknown BoucWen hysteresis model, and its mathematical expression is as follows:

[0018]

[0019] Among them, u(t) is the control input to be designed, and there exists a positive constant U M Make |u(t)|≤U M ; is the hysteresis part of the system; μ1 is a constant, α is the restoring force amplitude parameter, β and γ are the parameters of the shape and amplitude of the hysteresis respectively, and m is expressed as the smoothness from the initial slope to the asymptotic slope of the control hysteresis. The above parameters determine the shape and size of the hysteresis loop.

[0020] Wherein, the step S2 specifically includes the following steps:

[0021] Set f(x1,x2)=a1x1 cosx1+a2 cosx1+a3x2, g(x1)=a4μ1cosx1, At the same time, the unknown nonlinear function in the hysteresis nonlinear system Approximated by fuzzy logic system, that is for The estimated value of

[0022] set up in is the optimal weight vector, and are the fuzzy approximation error and the fuzzy minimum approximation error respectively, and there is an unknown positive constant Make and

[0023]

[0024] There exists a positive constant l iMake The unknown nonlinear function g(x1) is approximated by a fuzzy logic system; the following fuzzy state observer is designed:

[0025]

[0026] Among them, Λ1 and Λ2 are the state observer gains.

[0027] Wherein, the step S3 specifically includes the following steps:

[0028] First, introduce the coordinate transformation s1=x1-y r , Where s1 is the tracking error; s2 is the error surface; α2 is the virtual control function x 1,d Output through a first-order filter:

[0029] α2(0)=x 1,d (0)

[0030] Among them, τ1 is the time constant;

[0031] S3-1. Design virtual control law x 1,d for:

[0032]

[0033] Where c1>0;

[0034] S3-2. Design sliding surface s:

[0035]

[0036] in, and is the adaptive parameter, and is a normal number, p and q are both ratios of two positive odd numbers, and p>1, 0 <q<1;

[0037] Based on the equivalent control method, solve Design the following adaptive terminal sliding mode controller:

[0038]

[0039] Among them, the parameters to be designed κ1,λ1,λ2>0, is a small normal number, is the estimated value of λ2, and

[0040] The designed adaptive law is:

[0041]

[0042] Among them, r f , σ f , and are all normal numbers to be designed.

[0043] Wherein, the step S4 specifically includes the following steps:

[0044] The Lyapunov equation is selected as follows:

[0045]

[0046] Derivative of the Lyapunov equation yields:

[0047]

[0048] When the design parameter Λ is selected min ,κ1,λ1,σ f , σ g , and Let ρ>0. For any initial condition, suppose there exists a positive constant ζ such that V(0)≤ξ; let Then V(t)≤ζ(t≥0);

[0049] Integrate both sides of the derivative of the Lyapunov equation on [0, t]:

[0050]

[0051] It is found that all signals in the hysteresis nonlinear system are bounded, and both the observation error and the tracking error converge to a small neighborhood near the origin.

[0052] Beneficial effects:

[0053] 1. The present invention includes the establishment of a hysteresis nonlinear system model of a flexible joint rotation driver, and the design of a state observer and a terminal sliding mode controller. The research object is the angular displacement tracking control of a bionic robot flexible joint based on smart materials. The two performances of observation error and tracking error are selected, and then the state observer and the adaptive terminal sliding mode controller are designed to ensure that all signals in the closed-loop system are bounded, so as to achieve fast and accurate angular displacement tracking control of the hysteresis nonlinear system of the bionic robot flexible joint rotation driver and strong robustness of the system under specific working conditions.

[0054] 2. The present invention adopts the Bouc-Wen model that can separate hysteresis terms to characterize the hysteresis characteristics of the nonlinear system; uses a fuzzy logic system to approximate the unknown nonlinear function of the system, and on this basis further designs a fuzzy state observer to estimate the unmeasurable state of the system; based on the state observer and the hysteresis nonlinear system model of the bionic robot flexible joint rotation driver, the present invention combines sliding mode control theory, inversion technology and adaptive technology to design an adaptive terminal sliding mode controller that does not require feedforward inverse compensation; analyzes the rationality of the designed adaptive controller and the stability of the closed-loop system through Lyapunov stability theory.

[0055] 3. The present invention solves the problem that the state variables of existing hysteresis nonlinear systems are difficult to measure directly, simplifies the design process of the controller, reduces the calculation burden of the system, and realizes fast and accurate robust angular displacement tracking control of the bionic robot flexible joint rotation drive system under factors such as unmeasurable system states and external disturbances. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a schematic diagram of a flexible joint rotation driver in an embodiment of the present invention;

[0057] Figure 2 is a structural diagram of an adaptive terminal sliding mode control device for a hysteresis nonlinear system in an embodiment of the present invention;

[0058] Figure 3 The flowchart of the hysteresis nonlinear system tracking control method in the embodiment of the present invention. DETAILED DESCRIPTION

[0059] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0060] The present invention proposes a bionic robot flexible joint angular displacement tracking control method based on terminal sliding mode, including the establishment of a flexible joint driver hysteresis nonlinear system model, the design of a fuzzy state observer and an adaptive terminal sliding mode controller. The present invention adopts the Bouc-Wen model to characterize the hysteresis characteristics of the flexible joint driver system, reduces the difficulty of controller design, and alleviates the calculation burden of the drive system; by designing a state observer, the constraint problem that the flexible joint driver system platform has only a displacement sensor is solved; finally, based on the sliding mode control theory and adaptive technology, an adaptive terminal sliding mode controller without feedforward inverse compensation is designed. Specifically, the present invention selects the two properties of observation error and tracking error, and then designs a state observer and an adaptive terminal sliding mode controller to ensure that all signals in the closed-loop system are bounded, so as to realize the fast and accurate angular displacement tracking control of the hysteresis nonlinear system of the bionic robot flexible joint rotation driver under specific working conditions and the strong robustness of the system.

[0061] Figure 1 is a schematic diagram of a flexible joint rotation driver; the structure of its control device is as follows Figure 2 As shown, the tracking control method of the hysteresis nonlinear system of the flexible joint actuator mainly includes establishing a hysteresis nonlinear system model, constructing a fuzzy state observer and designing an adaptive terminal sliding mode controller.

[0062] Figure 3 This is a flow chart of the angular displacement tracking control method of the flexible joint drive system based on the adaptive terminal sliding mode of the present invention, including using the Bouc-Wen model to characterize the hysteresis characteristics of the system; designing a fuzzy state observer to estimate the unmeasurable system state variables; combining the inversion technology, sliding mode control theory and adaptive technology to design an adaptive terminal sliding mode controller that does not require feedforward inverse compensation; finally, using the Lyapunov stability theory to analyze the rationality of the designed adaptive terminal sliding mode controller and the stability of the closed-loop system.

[0063] The terminal sliding mode-based bionic robot flexible joint angular displacement tracking control method of the present invention specifically comprises the following steps:

[0064] S1. Based on the Bouc-Wen model to characterize the hysteresis characteristics of the system, a hysteresis nonlinear system model of the bionic robot flexible joint actuator is established, and its mathematical model is:

[0065]

[0066] Where x0 and x are the initial length and deformed length of the linear displacement actuator (AD), L0 and L are the initial length and deformed length of the spring (BC), and k1 is the spring constant. m, k, c represent mass, damping coefficient, and elastic coefficient respectively; I is the moment of inertia of AB about the midpoint O; w is the stress generated by voltage; is the fluid resistance torque.

[0067] Let x1 = θ, a2=C / I,a3=-ξ / I,a4=-l1 / I,Equation (1) can be rewritten as:

[0068]

[0069] Where d1 is the external disturbance of the system, w(u(t)) is the output of the unknown Bouc-Wen hysteresis model, and its mathematical expression is as follows:

[0070]

[0071] Among them, u(t) is the control input to be designed, and there exists a positive constant U M Make |u(t)|≤U M ; is the hysteresis part of the system; μ1 is a constant, α is the restoring force amplitude parameter, β and γ are the parameters of the shape and amplitude of the hysteresis respectively, and m is expressed as the smoothness from the initial slope to the asymptotic slope of the control hysteresis. The above parameters determine the shape and size of the hysteresis loop.

[0072] S2. Approximate the unknown nonlinear functions of the system through fuzzy logic system, design fuzzy state observer, and estimate the amount of information of unmeasurable system state.

[0073] S2-1. From equation (2), system (1) can be rewritten as:

[0074]

[0075] make f(x1,x2)=a1x1 cosx1+a2 cosx1+a3x2,g(x1)=a4μ1cosx1,Equation (4) can be rewritten as:

[0076]

[0077] In the present invention, the unknown nonlinear function in the hysteresis nonlinear system Approximated by fuzzy logic system, that is for The estimated value of Then, define, in is the optimal weight vector, and are the fuzzy approximation error and the fuzzy minimum approximation error respectively, and there is an unknown positive constant Make and In addition, there exists a positive constant l i Make Similarly, the unknown nonlinear function g(x1) is approximated by a fuzzy logic system.

[0078] definition System (5) can be rewritten as:

[0079]

[0080] In the formula, Assume there is an unknown positive constant Make

[0081] Design the following fuzzy state observer:

[0082]

[0083] Among them, Λ1 and Λ2 are the state observer gains.

[0084] From equations (6) and (7), we can get the error equation:

[0085]

[0086] S2-2. Select Lyapunov function Taking its derivative we get:

[0087]

[0088] According to Young's inequality, we can get:

[0089]

[0090] Among them, Λ min =min(Λ),

[0091] S3. Based on the fuzzy state observer, combined with the inversion technology, adaptive technology and sliding mode control theory, an adaptive terminal sliding mode controller is designed. The details are as follows:

[0092] S3-1. First, introduce the coordinate transformation s1 = x1-y r , Where s1 is the tracking error; s2 is the error surface; α2 is the virtual control function x 1,d Output through a first-order filter:

[0093]

[0094] Among them, τ1 is the time constant.

[0095] S3-2. Design virtual control law x 1,d for:

[0096]

[0097] Among them, c1>0.

[0098] S3-3. Design sliding surface s:

[0099]

[0100] in, and is the adaptive parameter, and is a normal number, p and q are both ratios of two positive odd numbers, and p>1, 0 <q<1。

[0101] In the designed integral terminal sliding surface, the tracking performance of the system is improved by introducing adaptive parameters m1 and m2. First, the fractional power function sig q (s i ) is designed in the integral terminal sliding surface to avoid singularity problems. When the system state is far from the equilibrium point, m1 tends to 1 and m2 tends to 0, which means that sig p (s i ) plays a major role, thereby accelerating the convergence of the driving system; conversely, when the system state approaches the equilibrium point, m1 tends to 0, and m2 tends to 1, then sig q (s i ) will continue to force the system state to converge to the equilibrium point, which improves the control accuracy and convergence speed of the closed-loop system while improving the system robustness and avoiding the chattering problem.

[0102] When considering system disturbances, mainly including uncertain factors such as model error, hysteresis nonlinearity and external disturbances, the upper limit of system disturbances is not easy to obtain, and a large switching gain will cause chattering. Based on the equivalent control method, the present invention solves Design the following adaptive terminal sliding mode controller:

[0103]

[0104] Among them, the parameters to be designed κ1,λ1,λ2>0, is a small normal number, is the estimated value of λ2, and

[0105] The designed adaptive law is:

[0106]

[0107] Among them, r f , σ f 、r g , σ g , and are all normal numbers to be designed.

[0108] S4. Use Lyapunov stability theory to analyze the rationality of the designed adaptive controller and the stability of the closed-loop system. The details are as follows:

[0109] Constructing Lyapunov functions in According to formula (15) and Young's inequality, The derivative is:

[0110]

[0111] Constructing Lyapunov functions According to equations (10), (16) and Young's inequality, taking the derivative of V yields:

[0112]

[0113] where λ max (P) is the maximum eigenvalue of matrix P.

[0114] When the design parameter Λ is selected min ,κ1,λ1,σ f , σ g , and Let ρ>0. For any initial condition, the present invention assumes that there exists a positive constant ζ such that V(0)≤ξ. Let Then V(t)≤ζ(t≥0).

[0115] Integrate both sides of equation (17) on [0, t]:

[0116]

[0117] According to formula (18), all signals in the hysteresis nonlinear system are bounded, and the observation error and tracking error can converge to a small neighborhood near the origin. This means the rationality of the adaptive terminal sliding mode controller designed by the present invention, and proves that the hysteresis nonlinear system of the bionic robot flexible joint rotation driver can still operate stably under the action of the designed controller when considering the influencing factors such as hysteresis nonlinearity and external disturbance.

[0118] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A bionic robot flexible joint angular displacement tracking control method based on terminal sliding mode, characterized in that: The tracking control method comprises the following steps: S1. Based on the BoucWen model to characterize the hysteresis characteristics of the system, a hysteresis nonlinear system model of the bionic robot flexible joint actuator is established; S2. Approximate the unknown nonlinear function of the system through the fuzzy logic system, design the fuzzy state observer, and estimate the amount of information of the unmeasurable system state; S3. Based on the fuzzy state observer, combined with the inversion technology, adaptive technology and sliding mode control theory, an adaptive terminal sliding mode controller is designed; S4. Use Lyapunov stability theory to analyze the rationality of the designed adaptive terminal sliding mode controller and the stability of the closed-loop system.

2. The method according to claim 1, characterized in that The mathematical model of the revolute joint drive system in step S1 is: Where x0 and x are the initial length and the length after deformation of the displacement actuator, respectively; L0 and L are the initial length and the length after deformation of the spring, respectively; k1 is the spring constant; m, k, c represent mass, damping coefficient, and elastic coefficient respectively; I is the moment of inertia of AB about the midpoint O; w is the stress generated by voltage; is the fluid resistance moment; Let x1 = θ, a2=C / I,a3=-ξ / I,a4=-l1 / I,the mathematical model is rewritten as: Where d1 is the external disturbance of the system, w(u(t)) is the output of the unknown BoucWen hysteresis model, and its mathematical expression is as follows: Among them, u(t) is the control input to be designed, and there exists a positive constant U M Make |u(t)|≤U M ; is the hysteresis part of the system; μ1 is a constant, α is the restoring force amplitude parameter, β and γ are the parameters of the shape and amplitude of the hysteresis respectively, and m is expressed as the smoothness from the initial slope to the asymptotic slope of the control hysteresis. The above parameters determine the shape and size of the hysteresis loop.

3. The method according to claim 2, characterized in that The step S2 specifically includes the following steps: Set f(x1,x2)=a1x1cosx1+a2cosx1+a3x2, g(x1)=a4μ1cosx1, At the same time, the unknown nonlinear function in the hysteresis nonlinear system Approximated by fuzzy logic system, that is for The estimated value of set up in is the optimal weight vector, and are the fuzzy approximation error and the fuzzy minimum approximation error respectively, and there is an unknown positive constant M θf Make and There exists a positive constant l i Make The unknown nonlinear function g(x1) is approximated by a fuzzy logic system; the following fuzzy state observer is designed: Among them, Λ1 and Λ2 are the state observer gains.

4. The method according to claim 3, characterized in that The step S3 specifically includes the following steps: First, introduce the coordinate transformation s1=x1-y r , Where s1 is the tracking error; s2 is the error surface; α2 is the virtual control function x 1,d Output through a first-order filter: Among them, τ1 is the time constant; S3-1. Design virtual control law x 1,d for: Where c1>0; S3-2. Design sliding surface s: in, and is the adaptive parameter, and is a normal number, p and q are both ratios of two positive odd numbers, and p>1, 0 <q<1; Based on the equivalent control method, solve Design the following adaptive terminal sliding mode controller: Among them, the parameters to be designed is a small normal number, is the estimated value of λ2, and The designed adaptive law is: Among them, r f , σ f , and are all normal numbers to be designed.

5. The method according to claim 3 or 4, characterized in that The step S4 specifically includes the following steps: The Lyapunov equation is selected as follows: Derivative of the Lyapunov equation yields: When the design parameter Λ is selected min ,κ1,λ1,σ f , σ g , and Let ρ>

0. For any initial condition, suppose there exists a positive constant ζ such that V(0)≤ξ; let Then V(t)≤ζ(t≥0); Integrate both sides of the derivative of the Lyapunov equation on [0, t]: It is found that all signals in the hysteresis nonlinear system are bounded, and both the observation error and the tracking error converge to a small neighborhood near the origin.

Citation Information

Cited By

  • Track control method based on time-varying fuzzy sliding mode

    CN120010561A