Robust adaptive control method for flexible manipulator with unknown dead zone
By establishing a dynamic model of the flexible robot arm and decomposing unknown dead zones, designing a robust adaptive controller, the stability and control performance problems of the flexible robot arm system in unknown dead zones are solved, and more efficient control effects are achieved. Cost reduction.
Patent Information
- Application Number
- CN202210240647.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-10
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2042-03-10
AI Technical Summary
The prior art is difficult to effectively control flexible robot arms with unknown dead zones, resulting in system instability and degradation of control performance, especially when facing nonlinear characteristics such as saturation and dead zones, traditional methods such as modal control and distributed control exist. limitation.
By establishing a dynamic model of the flexible robot arm, decompose unknown dead zones into linear and nonlinear parts, design a robust adaptive controller, estimate dead zone parameters using adaptive methods, and construct Lyapunov function to analyze system stability, combined with MATLAB Digital simulation optimizes controller parameters.
More stable and precise tracking control of flexible robot arms with unknown dead zones is achieved, which suppresses elastic vibration of the robot arms, reduces the demand for sensors and controllers, has low cost and significant control effect.
Smart Images

Figure CN114740714B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of automatic control technology, in particular to a robust adaptive control method for a flexible mechanical arm with unknown dead zone. Background Art
[0002] With the breakthrough of the new generation of robot technology, robots are no longer only used in the industrial field. Robots, especially robotic arms, are also used in the fields of ocean, construction, aerospace, etc. to carry out related work. Faced with an increasingly complex external environment, rigid robotic arms that are bulky, inflexible, and have a single function can no longer meet the needs. Compared with traditional rigid robotic arms, flexible robotic arms have the advantages of low energy consumption, high speed, light weight, and high precision, so they are widely used in various fields. However, due to the flexibility of its own materials, the robotic arm is easily affected by external interference. In the actual working process, it will produce unnecessary mechanical vibrations, affecting the operating performance of the system; in order to solve its vibration problem, a large number of scholars and experts have conducted in-depth research and proposed a series of control methods: including modal control, distributed control, boundary control, etc. Modal control only considers some key modes of the system, ignores other modes, and approximates the dynamic model of the distributed parameter system with infinite-dimensional characteristics to a finite-dimensional system model. This control method can suppress vibration to a certain extent, but it is easy to cause spillover effects due to ignoring high-frequency modes. Distributed control has a better effect in suppressing vibration than modal control, but it requires a large number of sensors and actuators, and the possibility of implementation in practical applications is relatively small. The boundary control method only requires the installation of sensors and controllers at both ends of the system, and is considered to be the most effective control method. For example, patent number CN201610496241.9 provides a vibration boundary control method for a flexible manipulator with output constraints; patent number CN201711499692.9 proposes a boundary control method for anti-saturation of a flexible Timoshenko beam manipulator, which can not only ensure the operating performance of the system, but also deal with the problem of input saturation. The above inventions are all for flexible manipulators with nonlinear characteristics. However, for flexible manipulators with unknown dead zones, the above solutions will have great limitations; in industrial control devices, nonlinear characteristics are inevitable, such as saturation, dead zones, hysteresis, etc. The input dead zone is usually caused by the constraints of the mechanical structure and technical conditions of the system. If the input dead zone is ignored in the control design, it will affect the control performance of the system and even cause system instability. Summary of the invention
[0003] In view of the shortcomings of the prior art, the present invention provides a robust adaptive control method for a flexible robotic arm with unknown dead zone, which has the advantages of decomposing the unknown dead zone into linear and nonlinear parts, applying an adaptive method to estimate the parameters of the unknown dead zone, and designing a robust adaptive controller that can compensate for the influence of the dead zone, thereby solving the problem of more stable and precise tracking and control of a flexible robotic arm with an unknown dead zone.
[0004] In order to achieve the above-mentioned purpose of more stable and accurate tracking and control of a flexible manipulator with an unknown dead zone, the present invention provides the following technical solution: a robust adaptive control method for a flexible manipulator with an unknown dead zone, comprising the following steps:
[0005] 1) Establish a dynamic model of the system
[0006] For the sake of simplicity, some formulas are simplified as follows:
[0007] ,
[0008] ,
[0009] The kinetic energy of the single-link flexible robotic arm system is expressed as:
[0010] ,
[0011] The potential energy of the flexible robotic arm system is expressed as:
[0012]
[0013] Where EI represents the bending stiffness of the flexible manipulator and T is the constant tension of the manipulator.
[0014] System damping, the total virtual work done by the controller on the flexible robotic arm system:
[0015] ,
[0016] The kinetic energy of the system Potential Energy , the virtual work done by the controller Substituting the Hamilton principle, the dynamic model of the flexible robotic arm system is as follows:
[0017]
[0018]
[0019]
[0020]
[0021]
[0022] , It is the control signal before passing through the dead zone link;
[0023] 2) Unknown dead zone in the system
[0024] ,
[0025] By converting it, we can get
[0026]
[0027] in, For the subsequent controller calculation, we define two parameters here, namely and The upper bound of ;
[0028] 3) Building a robust adaptive controller
[0029]
[0030]
[0031] Among them, the observed variables The system parameters , , dead zone slope and dead zone width are estimated; is an auxiliary system, and
[0032]
[0033] The observed variables are iterated through an adaptive method
[0034]
[0035]
[0036]
[0037] ;
[0038] 4) Construct Lyapunov function And analyze the stability of the flexible robotic arm system, define for:
[0039]
[0040] The content of each sub-formula is as follows
[0041] , represents the energy term;
[0042] , indicating auxiliary items;
[0043] , , represents the cross term;
[0044] , represents additional items,
[0045] in, The new variable in the additional term is the error term corresponding to the observed variable, defined as
[0046]
[0047] First verify the Lyapunov function The boundedness of The negative definiteness of , it is concluded that the system is asymptotically stable;
[0048] 5) Use MATLAB to digitally simulate the system
[0049] Use MATLAB simulation software to digitally simulate the flexible robotic arm system, analyze the simulation results and determine whether the control effect meets the requirements. If it does not meet the requirements, modify the gain parameters of the boundary controller. ; If the requirements are met, then end.
[0050] Preferably, in step 1 represents the various positions of the flexible robotic arm, Indicates the system running time. represents the moment of inertia of the wheel hub, L is the length of the single-link flexible robotic arm, Represents the uniform mass per unit length of the flexible manipulator and the absolute displacement of the manipulator in xoy coordinates Defined as ,in represents the elastic deformation of the flexible manipulator system at time t and position z in the xoy coordinate system, Indicates the rotation angle of the robot arm.
[0051] Preferably, in step 1 is the variation symbol, is the damping coefficient of the system, is the system input.
[0052] Preferably, in step 2 is the slope of the input dead-band link, and is the range of the dead zone and satisfies and , these three parameters are unknown constants.
[0053] Preferably, in step 3 is the gain parameter, and its value is greater than 0.
[0054] Preferably, in step 3 is the gain parameter, and its value is greater than 0.
[0055] Preferably, verifying the stability of the system includes the following two steps:
[0056] (2.1) Verify the Lyapunov function The boundedness of is proved as follows:
[0057] right Scaling the third term of
[0058] ,
[0059] in, is a positive constant and satisfies .
[0060] Then the following formula holds
[0061]
[0062] To satisfy
[0063] ,
[0064] in, ,
[0065] That is, the Lyapunov function The boundedness of is verified.
[0066] Equation Section (Next) (2.2) Verification Negative
[0067] First derivative with respect to time
[0068] (15)
[0069] Substituting formula (1) into formula (15), we obtain
[0070] (16)
[0071] Combining equation (2) and equation (3), we can obtain
[0072] (17)
[0073] First derivative with respect to time
[0074] (18)
[0075] Substituting into equations (4), (6)-(8), we can obtain
[0076] (19)
[0077] Simplifying and scaling equation (19), we can get
[0078] (20)
[0079] First derivative with respect to time
[0080] (twenty one)
[0081] Similar to the first term, substituting equation (1) into the equation, we can get
[0082] (twenty two)
[0083] By integrating it by parts and scaling it, we can get
[0084] (twenty three)
[0085] in, ,
[0086] First derivative with respect to time
[0087] (twenty four)
[0088] Substituting into equations (10)-(13), we can get
[0089] (25)
[0090] Substituting equations (17), (20), (23), and (25) into ,get
[0091] (26)
[0092] in, ,parameter Satisfying conditions (27)-(35):
[0093] (27)
[0094] (28)
[0095] (29)
[0096] (30)
[0097] (31)
[0098] (32)
[0099] (33)
[0100] (34)
[0101] (35)
[0102] Combination The conclusion of boundedness is that
[0103] (36)
[0104] in, .
[0105] Right now The negative positivity of is verified.
[0106] Multiply both sides of equation (36) by We get formula (37)
[0107] (37)
[0108] Integrating both sides of equation (37) yields
[0109] (38)
[0110] Depend on There is boundedness, so we can get
[0111] (39)
[0112] Substituting into equation (38), equation (39) can be written as
[0113] (40)
[0114] Therefore, we can get equations (41)-(43)
[0115] (41)
[0116] (42)
[0117] (43).
[0118] Preferably, from equations (41) and (43), we can obtain that the vibration amount and deflection angle of the flexible robotic arm can achieve final consistent stability and boundedness, indicating that the designed control method can effectively compensate for the dead zone influence and ensure the working accuracy of the system.
[0119] Compared with the prior art, the present invention provides a robust adaptive control method for a flexible manipulator with unknown dead zone, which has the following beneficial effects:
[0120] 1. This method of robust adaptive control of a flexible manipulator with unknown dead zone is based on the design of a robust adaptive controller based on the boundary control algorithm. The controller can achieve control by simply installing it at the end of the flexible manipulator. The controller can not only suppress the elastic vibration of the flexible manipulator and achieve tracking performance, but also compensate for the influence of the unknown dead zone on the input of the system control signal.
[0121] 2. The robust adaptive control method for the flexible manipulator with unknown dead zone adopts an adaptive method, so that the control algorithm does not need to clarify the system parameters in advance, but can be gradually iterated through the adaptive law.
[0122] 3. This robust adaptive control method for flexible robotic arms with unknown dead zones requires fewer sensors and controls, has low cost, and has broad prospects. BRIEF DESCRIPTION OF THE DRAWINGS
[0123] Figure 1 This is a schematic diagram of the operation of the flexible robotic arm;
[0124] Figure 2 It is an implementation flow chart of the robust adaptive control method for flexible manipulator vibration with unknown dead zone of the present invention;
[0125] Figure 3 is the elastic deformation of the flexible manipulator without control ;
[0126] Figure 4 is the angular position of the hub of the flexible manipulator without control ;
[0127] Figure 5 is the vibration of the flexible manipulator without control ;
[0128] Figure 6is the elastic deformation of the flexible manipulator after control is applied ;
[0129] Figure 7 is the angular position of the hub of the flexible manipulator after control is applied ;
[0130] Figure 8 is the vibration amount of the flexible manipulator after control is applied ;
[0131] Fig. 9 is the angular error of the hub of the flexible manipulator after control is applied ;
[0132] Fig.10 Robust adaptive control for design . DETAILED DESCRIPTION
[0133] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0134] Example:
[0135] Figure 2 is a flowchart of the implementation of the robust adaptive control method of the flexible manipulator for unknown dead zones of the present invention. Figure 2 As shown, the present invention is a robust adaptive control method for a flexible manipulator with unknown dead zone, characterized by comprising the following steps: ST1, establishing a dynamic model of the system. For the convenience of expression, some formulas are simplified as follows: ,
[0136] .
[0137] The kinetic energy of the flexible robotic arm system is expressed as:
[0138] in, represents the various positions of the flexible robotic arm, Indicates the system running time. represents the moment of inertia of the wheel hub, L is the length of the single-link flexible robotic arm, Represents the uniform mass per unit length of the flexible manipulator and the absolute displacement of the manipulator in xoy coordinates Defined as ,in represents the elastic deformation of the flexible manipulator system at time t and position z in the xoy coordinate system, Represents the rotation angle of the robot arm. The potential energy of the flexible robot arm system is expressed as:
[0139] Where EI represents the bending stiffness of the flexible manipulator, and T is the constant tension of the manipulator. System damping, the total virtual work done by the controller on the flexible manipulator system:
[0140] in, is the variation symbol, is the damping coefficient of the system, is the system input. Potential Energy , the virtual work done by the controller Substituting the Hamilton principle, the dynamic model of the flexible robotic arm system is as follows:
[0141]
[0142]
[0143]
[0144]
[0145] , It is the control signal before passing through the dead zone link.
[0146] ST2. Unknown dead zone in the system
[0147] in, is the slope of the input dead-band link, and is the range of the dead zone and satisfies and , these three parameters are unknown constants. By converting them, we can get in, For the subsequent controller calculation, we define two parameters here, namely and The upper bound of ST3. Constructing a robust adaptive controller in, is the gain parameter, and its value is greater than 0; the observed variable The system parameters , , dead zone slope and dead zone width are estimated; is an auxiliary system, and The observed variables are iterated through an adaptive method in, is the gain parameter, and its value is greater than 0. ST4. Construct Lyapunov function And analyze the stability of the flexible manipulator system and define the Lyapunov function of the flexible manipulator system The content of each sub-formula is as follows: , represents the energy term; , indicating auxiliary items; , represents the cross term; , represents an additional term. Among them, The new variable in the additional term is the error term corresponding to the observed variable, defined as . First verify the Lyapunov function The boundedness of The negative definiteness of , shows that the system is asymptotically stable. Verify the Lyapunov function The boundedness of is proved as follows: Scaling the third term of in, is a positive constant and satisfies . Then the following formula holds To satisfy in, . That is, the Lyapunov function The boundedness of is verified. Negative First derivative with respect to time (a) Substituting equation (1) into equation (a) yields (b) Partially integrate equation (b) and combine equations (2) and (3) to obtain: (c) First derivative with respect to time (d) Substituting into equations (4), (6)-(8), we can obtain (e) Simplifying and scaling equation (e), we can get (f) First derivative with respect to time (g) Similar to the first term, substituting equation (1) into the equation, we can obtain (h) By integrating it by parts and scaling it, we can get (i) Of which, . First derivative with respect to time (j) Substituting into equations (10)-(13), we can obtain (k) Substitute equations (17), (20), (23), and (25) into ,get (l) Among them, ,parameter The following conditions are met: , , , , , . Combined The conclusion of boundedness is that (m) Of which, .Right now The negative definiteness of is verified by multiplying both sides of equation (m) by The formula (n) is obtained (n) Integrating both sides of equation (n) yields (o) By There is boundedness, so we can get Substituting (p) into formula (o), formula (p) can be written as (q) Therefore, we can get the formula (r)-(t) (r) (s) (t) From equation (r) and equation (t), we can get that the vibration amount and deflection angle of the flexible robotic arm can achieve final consistent stability and boundedness, indicating that the designed control method can effectively compensate for the dead zone effect and ensure the working accuracy of the system. ST5. Simulate the flexible robotic arm system on the MATLAB platform. ST6. View and analyze the simulation effect of the flexible robotic arm. Figure 3 , Figure 4 and Figure 5 The following are the schematic diagrams of elastic deformation, angle position and vibration of the flexible manipulator when no control is applied. As shown in the figure, when no control is applied, vibrations occur in various parts of the manipulator, and the angle position also oscillates periodically and cannot be stabilized within a certain range. Figure 6 , Figure 7 and Figure 8 Schematic diagram of vibration simulation when adding the control of the present invention. Figure 6 and Figure 7 As shown, the present invention is used to suppress the vibration of the flexible robotic arm. After t=1.2s, the amplitude of the flexible robotic arm tends to be relatively stable, and the angle deflection can also track to the vicinity of the expected trajectory. Fig. 9 It is a schematic diagram of increasing the angle tracking error of the control system of the present invention, which can be stabilized near 0 after 1.2s. Fig.10 The signal of the designed controller is shown. In summary, the designed controller can eliminate the adverse effects of the dead zone on the system, and at the same time achieve vibration suppression and trajectory tracking of the system. ST6. According to the simulation effect of step ST5, determine whether the gain parameters need to be adjusted. According to the simulation results, determine whether the elastic deformation suppression and angle tracking of the flexible manipulator meet the requirements. If not, readjust the gain parameters of the boundary controller. ; If the requirements are met, then end. The above specific examples are only used as examples to provide a detailed and exemplary description of one form of the present invention. Without departing from the principle of the present invention, any technical solution obtained by a person skilled in the art based on the above examples through modification, equivalent replacement, etc., is within the protection scope of the present invention.
[0148] The present invention is a robust adaptive control method for a flexible manipulator with unknown dead zones. The method analyzes the dynamics of a single-link flexible manipulator system and establishes a model, decomposes the unknown dead zones therein, estimates the unknown parameters in the system by an adaptive method, and designs a robust adaptive controller capable of compensating for the dead zone in combination with the dynamic model. A Lyapunov function is then constructed to perform stability analysis on the flexible manipulator with unknown dead zones, and MATLAB is used to perform numerical simulation on the motion state of the system. The design parameters of the control system are adjusted according to the simulation results so as to eliminate the dead zone effects, effectively suppress the vibration of the flexible manipulator during operation, and realize angle tracking performance.
[0149] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A robust adaptive control method for a flexible manipulator with unknown dead zone, characterized in that: The following steps are involved: 1) Establish a dynamic model of the system For the sake of simplicity, some formulas are simplified as follows: , , The kinetic energy of the single-link flexible robotic arm system is expressed as: , The potential energy of the flexible robotic arm system is expressed as: Where EI represents the bending stiffness of the flexible manipulator, T is the constant tension of the manipulator, System damping, the total virtual work done by the controller on the flexible robotic arm system: , The kinetic energy of the system Potential Energy , the virtual work done by the controller Substituting the Hamilton principle, the dynamic model of the flexible robotic arm system is as follows: , It is the control signal before passing through the dead zone link; 2) Unknown dead zone in the system , By converting it, we can get in, For the subsequent controller calculation, we define two parameters here, namely and The upper bound of ; 3) Building a robust adaptive controller Among them, the observed variables The system parameters , , dead zone slope and dead zone width are estimated; is an auxiliary system, and The observed variables are iterated through an adaptive method ; 4) Construct Lyapunov function And analyze the stability of the flexible robotic arm system, define for: The content of each sub-formula is as follows , represents the energy term; , indicating auxiliary items; , , represents the cross term; , represents additional items, in, The new variable in the additional term is the error term corresponding to the observed variable, defined as First verify the Lyapunov function The boundedness of The negative definiteness of , it is concluded that the system is asymptotically stable; 5) Use MATLAB to digitally simulate the system Use MATLAB simulation software to digitally simulate the flexible robotic arm system, analyze the simulation results and determine whether the control effect meets the requirements. If not, modify the gain parameters of the boundary controller. ; If the requirements are met, then end.
2. The robust adaptive control method for a flexible manipulator with unknown dead zone according to claim 1, characterized in that: In step 1 represents the various positions of the flexible robotic arm, Indicates the system running time. represents the moment of inertia of the wheel hub, L is the length of the single-link flexible robotic arm, Represents the uniform mass per unit length of the flexible manipulator and the absolute displacement of the manipulator in xoy coordinates Defined as ,in represents the elastic deformation of the flexible manipulator system at time t and position z in the xoy coordinate system, Indicates the rotation angle of the robot arm.
3. The robust adaptive control method for a flexible manipulator with unknown dead zone according to claim 1, characterized in that: In step 1 is the variation symbol, is the damping coefficient of the system, is the system input.
4. The robust adaptive control method for a flexible manipulator with unknown dead zone according to claim 1, characterized in that: In step 2 is the slope of the input dead-band link, and is the range of the dead zone and satisfies and , these three parameters are unknown constants.
5. The robust adaptive control method for a flexible manipulator with unknown dead zone according to claim 1, characterized in that: In step 3 is the gain parameter, and its value is greater than 0.
6. The robust adaptive control method for a flexible manipulator with unknown dead zone according to claim 1, characterized in that: In step 3 is the gain parameter, and its value is greater than 0.
Citation Information
Patent Citations
Vibration control method for flexible mechanical arm with output constraint
CN106078742A
Boundary control method for anti-saturation of flexible Timoshenko beam manipulator
CN108181836A
Robust sliding mode adaptive control method for manipulator system
CN108983606A
Combined sliding mode motion control method of flexible mechanical arm
CN111948943A
Robust high-performance control for robotic manipulators
US5049796A
