Fuzzy adaptive control method and system for uncertain robot system constraints

Abstract

The invention belongs to the technical field of robot control, and discloses a fuzzy adaptive control method and system for uncertain robot system constraints. A dead zone model is integrated into an uncertain robot kinetic equation, a robot system module is established, feedback joint displacement state information is output in real time through the robot system module, and unmeasurable speed and acceleration state information is processed through a filter module, a virtual controller module and a self-adaptive law module. And a virtual control signal and a fuzzy adaptive law parameter are obtained, and dynamic change information is obtained through a fuzzy logic system in a fuzzy adaptive controller module. According to the method, parameter uncertainty caused by unknown robot system parameters or uncertainty and dead zone characteristics caused by non-controlled object factors such as external disturbance are solved, and accurate tracking control of the system state under boundary constraint is realized.

Description

Technical Field

[0001] This invention relates to the field of robot control technology, and specifically to a fuzzy adaptive control method and system for uncertain robot system constraints. Background Technology

[0002] Industrial robots, with their advantages of high efficiency, high reliability, and strong adaptability, have been widely used in the manufacturing field and have become an indispensable key equipment in modern industrial systems. However, existing control theories and technologies are still insufficient in addressing many challenges faced in practical applications, such as high-precision trajectory tracking, input dead zone compensation, safety constraint control, and stability in complex environments. Among these, the dead zone problem is a particularly prominent and unavoidable engineering challenge in robot control. The actuator dead zone that is prevalent in robot systems causes control commands to be unable to be effectively responded to within a certain range, leading to phenomena such as insensitivity, lag in response, or even failure to complete tasks smoothly in specific positions or postures. This type of dead zone characteristic not only significantly reduces the control accuracy and dynamic performance of robots but may also cause serious consequences such as system oscillation or instability. Therefore, effective compensation and suppression of the dead zone effect are crucial.

[0003] Furthermore, due to production safety, workspace limitations, and obstacle interference, robots must meet strict state constraints during movement, such as joint angles, speeds, and safety boundaries of their workspace. Under complex tasks and multi-source interference conditions, a lack of reasonable constraint handling mechanisms can not only degrade system performance but also lead to execution failures or even operational hazards. Therefore, achieving effective state constraint control of robot systems is a crucial step in ensuring their safety and reliability.

[0004] Meanwhile, due to the incompleteness of existing theoretical frameworks and the lack of general-purpose control tools that can be directly applied to industrial settings, various external disturbances can still lead to unpredictable operational risks for uncertain robotic systems with dead-zone characteristics and physical constraints, making it difficult to effectively guarantee system performance. Therefore, developing adaptive control methods that balance robustness, safety, and high-performance control has become a crucial research direction urgently needing breakthroughs in the field of industrial robotics. Summary of the Invention

[0005] This invention proposes a fuzzy adaptive control method and system for uncertain robot system constraints to overcome the above-mentioned problems.

[0006] The technical solution of the present invention is as follows: A fuzzy adaptive control method for uncertain robot system constraints includes the following steps:

[0007] Step 1: Integrate the dead zone model into the uncertain robot dynamics equations to establish a robot system module; construct time-varying constraints and error constraints for the robot system module to limit joint displacement, joint velocity, tracking error, and virtual error surface; the robot system module obtains the output signal of the fuzzy adaptive controller module, calculates the control torque of the robot system module's actuator, and adjusts the robot's joint displacement according to the control torque;

[0008] Step 2: The filter module obtains the robot's joint displacement, velocity, and acceleration output by the robot system module, and calculates the output signal of the first-order filter constructed based on the virtual error surface; the virtual controller module obtains the output signal of the first-order filter of the filter module, and calculates the virtual control signal for fuzzy control.

[0009] Step 3: The fuzzy adaptive law module acquires the virtual control signal of the fuzzy control and calculates the dynamic change information of the fuzzy adaptive law module;

[0010] Step 4: The fuzzy adaptive controller module acquires dynamic change information and calculates the output signal of the fuzzy adaptive controller module.

[0011] The robot dynamics equations for the uncertainty mentioned in step 1 are as follows:

[0012]

[0013] in, , , These represent joint displacement, joint velocity, and joint acceleration, respectively. Represents a positive definite symmetric inertial matrix. Represents the unknown Coriolis force-centrifugal force matrix. The gravity vector matrix, It is an unknown Jacobian matrix, subject to external disturbances. , It is the upper bound of the disturbance. It is a control signal;

[0014] The dead zone model is as follows:

[0015]

[0016] in, Represents the input to the control law. The right slope represents the dead zone feature. The left slope represents the dead zone feature. Represents the right intercept point. Represents the left intercept point. It is the actuator control torque at time t;

[0017] A smooth and continuous mathematical model is used to perform an inverse transformation of the feedback linearization dead zone, and the dead zone model is transformed into the following form:

[0018]

[0019] In the formula, ,

[0020] in, This represents the linear part of the dead zone model after inverse transformation. This represents the nonlinear part of the dead-zone model after inverse transformation. Bounded, Represents the upper bound of the cutoff point;

[0021] Define state variables , Substituting the inversely transformed dead-zone model into the uncertain robot dynamics equations, we obtain the following expression for the robot system module:

[0022]

[0023] in, , This represents the displacement of the i-th joint, where i is from 1 to n; , Represents the velocity of the i-th joint; It is the control torque of the actuator of the robot system module; Represents external disturbances related to time; Represents the state variable The gravity vector matrix, It is the output feedback variable;

[0024] The time-varying constraint condition is: , ;in, and It is a continuous, differentiable, time-varying constraint bound, designed based on the physical limitations of the robot, and used to characterize safety requirements that change over time;

[0025] The robot system module assumes the existence of constants. and Satisfying the inequality and , ; and The i-th joint represents a continuous, differentiable, time-varying constraint bound; the robot system module introduces a function vector. and And satisfy the inequality and And the function components and , make the inequality and Established, Represents the expected displacement trajectory variable. The derivative of the expected displacement trajectory variable. ;

[0026] The error constraints of the robot system module include tracking error constraints and virtual error surface constraints:

[0027]

[0028] in, This represents the tracking error constraint. It is the tracking error obtained from real-time measurement. Represents the output and state variables of a first-order filter. deviation, This represents the virtual error surface constraint condition.

[0029] The filter module is established as follows:

[0030] by Based on this, a virtual error surface is established, Mapping to a new coordinate space; introducing a first-order filter to solve for the virtual control signal of fuzzy control;

[0031] The virtual error surface is as follows:

[0032]

[0033]

[0034]

[0035] in, It is the expected trajectory of joint movement. It is the output error of the first-order filter; The tracking error is the first One portion, It is the virtual error surface. One portion, The output error of the first-order filter is the... One component; Virtual control signals representing fuzzy control;

[0036] This represents the output of the virtual control signal after passing through a first-order filter.

[0037] The first-order filter is as follows:

[0038]

[0039] in, It is the time constant in the filter module;

[0040] The virtual controller module is established as follows;

[0041] The virtual controller module and its component form are as follows:

[0042]

[0043]

[0044] In the formula, This represents the constraint boundary repulsion term in the virtual controller module. , The tracking error constraint is the first One portion, The i-th component of the virtual control signal represents the fuzzy control signal. , ;in, This refers to the feedback gain parameter in the virtual controller module. It is a time-varying gain parameter. make sure Bounded, satisfying the inequality , .

[0045] The fuzzy adaptive law module is obtained based on the fuzzy logic system;

[0046] The fuzzy logic system is specifically as follows: It is a continuous function defined on a closed set. Above, using fuzzy logic systems express Satisfies the expression:

[0047]

[0048] in, Represents the fuzzy fundamental function vector. Indicates the upper limit. It is the optimal approximation error. Represents the ideal weight. Represents the total number of fuzzy rules, the th fuzzy fundamental function vectors Definition:

[0049]

[0050] It is a fuzzy membership function. The input vector representing the fuzzy membership function. The total number of input variables in the fuzzy logic system;

[0051] Based on the approximation properties of fuzzy logic systems, this method handles unknown nonlinear functions. The specific form of the components is as follows:

[0052]

[0053] In the formula, The positive definite symmetric inertial matrix is ​​the first... One portion, For external disturbances One portion, The Coriolis force-centrifugal force matrix is ​​the first... One portion, For the Jacobian matrix, the first... One portion, The gravity vector matrix is ​​the first One portion, For the ideal weights A vector, , The optimal approximation error is represented by the first... Each component is bounded. , This represents the upper bound of the optimal approximation error. This represents the square of the norm of the weight vector. Represents the optimal parameters The estimated value;

[0054] The mathematical model of the fuzzy adaptive law module is as follows:

[0055]

[0056] In the formula, , representing the constraint boundary repulsion term in the fuzzy adaptive law. The virtual error surface constraint condition is the first One portion, , These are the design parameters in the fuzzy adaptive law. These are the adaptive parameters of the fuzzy adaptive law; It is the dynamic change information of the fuzzy adaptive law parameter module, which will The input is fed into the fuzzy adaptive controller module.

[0057] The mathematical model of the fuzzy adaptive controller module is as follows:

[0058]

[0059] No. The components are:

[0060]

[0061] In the formula, Represents virtual error margin and The coupling factor corresponding to the gradient of the barrier potential field. It is the first One coupling factor component, The first part representing the linear part of the dead zone model after inverse transformation One portion, , , The feedback gain parameter in the fuzzy adaptive controller. The first-order filter represents the output of the virtual control signal. One derivative component, It is a time-varying gain parameter. make sure Bounded, satisfying the inequality , ;

[0062] The output signal of the fuzzy adaptive controller module is input to the robot system module to calculate the control torque of the actuator of the robot system module.

[0063] A fuzzy adaptive control system for uncertain robot system constraints, comprising a fuzzy adaptive control method for uncertain robot system constraints, including:

[0064] Robot system modules containing dead zone characteristics and uncertainties, filter modules, virtual controller modules, fuzzy adaptive law modules, and fuzzy adaptive controller modules;

[0065] The input of the filter module is connected to the output of the robot system module containing dead-zone characteristics and uncertainties; the output of the filter module is connected to the input of the virtual controller module; the output of the virtual controller module is connected to the input of the fuzzy adaptive law module; the output of the fuzzy adaptive law module is connected to the inputs of the virtual controller module and the fuzzy adaptive controller module respectively; the output of the fuzzy adaptive controller module is connected to the input of the robot system module containing dead-zone characteristics and uncertainties.

[0066] The fuzzy adaptive law module is used to calculate the dynamic change information of the adaptive law parameters and send the dynamic change information to the fuzzy adaptive controller module.

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] This invention acquires real-time feedback of joint displacements and obtains joint velocities and accelerations from robot system modules. It then calculates the output signal of a filter module, uses this output information to generate a virtual controller signal, and employs a fuzzy adaptive law module to identify adaptive law parameters online. Finally, it uses the control signal returned to the robot system module by the fuzzy adaptive controller module to calculate the actuator control torque. This control torque is then used to adjust joint displacements, enabling precise robot control. This achieves fuzzy adaptive state feedback control under the boundary constraints of uncertain robot systems with dead-zone characteristics.

[0069] This invention features high precision, robustness, and adaptability. Through an inverse transformation mechanism for feedback linearization dead zones, it effectively compensates for the impact of dead zone problems. The virtual control signal is processed by a first-order filter, effectively solving the derivative explosion problem of the virtual controller. The introduction of a fuzzy logic system, using the fuzziness of rules to encompass uncertainties in the robot system, maintains the stability of the approximation accuracy without frequent model recalibration. Its computational efficiency is far higher than complex deep learning models, meeting the requirements for real-time robot control. Applying a fuzzy adaptive controller to an uncertain robot system enables online identification of adaptive law parameters, precise control of joint displacement and velocity within time-varying constraints, and excellent tracking performance, ensuring stable operation throughout the closed-loop process. Attached Figure Description

[0070] Figure 1 This is a schematic diagram of the fuzzy adaptive state feedback control of an uncertain robot system with dead zone characteristics according to the present invention.

[0071] Figure 2 The diagram shows the tracking trajectory of joint displacement and reference signal under boundary constraints according to the present invention; (a) shows the joint displacement. Tracking trajectory, (b) is the joint displacement Tracking trajectory;

[0072] Figure 3 This is a schematic diagram of joint velocities under boundary constraints according to the present invention; (a) shows the joint velocities. (b) represents the joint velocity. ;

[0073] Figure 4 This is a schematic diagram of the robot adaptive controller signal of the present invention; (a) is the adaptive controller signal. (b) is the adaptive controller signal. ;

[0074] Figure 5This is a schematic diagram of the response curves of the adaptive law parameters of the present invention; (a) is the adaptive law parameters. (b) represents the adaptive law parameters. ;

[0075] Figure 6 (a) is the robot error phase diagram of the present invention; (b) is the tracking error phase diagram. Detailed Implementation

[0076] The present invention will be further described below with reference to the accompanying drawings and embodiments;

[0077] The robot system module is used to acquire the output signal of the fuzzy adaptive controller module. And according to the control torque Adjusting the joint displacement of the robot system module ;

[0078] The filter module is used to acquire the robot's joint displacements. Joint velocity and joint acceleration And calculate the output of the first-order filter based on the virtual error surface. ;

[0079] The virtual controller module is used to obtain the output of the first-order filter of the filter module. Data information of adaptive parameter estimates for fuzzy control According to the output of the first-order filter of the filter module Data information of adaptive parameter estimates for fuzzy control Calculate the virtual controller signal for fuzzy control ;

[0080] The fuzzy adaptive law module is used to acquire the virtual controller signal of the fuzzy control. And based on the virtual controller signal of the fuzzy control Data information for calculating the parameter estimates of the fuzzy adaptive law ;

[0081] The fuzzy adaptive controller module is used to acquire data information of the fuzzy adaptive law parameter estimates of the fuzzy control. And based on the estimated values ​​of the fuzzy adaptive law parameters of the fuzzy control. Calculate the output signal of the fuzzy adaptive controller .

[0082] The robot dynamics equations for the uncertainty mentioned in step 1 are as follows:

[0083]

[0084] in, , , These represent joint displacement, joint velocity, and joint acceleration, respectively. Represents a positive definite symmetric inertial matrix. Represents the unknown Coriolis force-centrifugal force matrix. The gravity vector matrix, It is an unknown Jacobian matrix, subject to external disturbances. , It is the upper bound of the disturbance. It is a control signal;

[0085] The dead zone model is as follows:

[0086]

[0087] in, Represents the input to the control law. The right slope represents the dead zone feature. The left slope represents the dead zone feature. Represents the right intercept point. Represents the left intercept point; It is the actuator control torque at time t;

[0088] A smooth and continuous mathematical model is used to perform an inverse transformation of the feedback linearization dead zone, and the dead zone model is transformed into the following form:

[0089]

[0090] In the formula, ,

[0091] in, This represents the linear part of the dead zone model after inverse transformation. This represents the nonlinear part of the dead-zone model after inverse transformation. Bounded, Represents the upper bound of the cutoff point;

[0092] Define state variables , Substituting the inversely transformed dead-zone model into the uncertain robot dynamics equations, we obtain the following expression for the robot system module:

[0093]

[0094] in, , This represents the displacement of the i-th joint, where i is from 1 to n; , Represents the velocity of the i-th joint; It is the control torque of the actuator of the robot system module; Represents external disturbances related to time; Represents the state variable The gravity vector matrix, It is the output feedback variable;

[0095] The time-varying constraint condition is: , ;in, and It is a continuous, differentiable, time-varying constraint bound, designed based on the physical limitations of the robot, and used to characterize safety requirements that change over time;

[0096] The robot system module assumes the existence of constants. and Satisfying the inequality and , ; and The i-th joint represents a continuous, differentiable, time-varying constraint bound; the robot system module introduces a function vector. and And satisfy the inequality and And the function components and , make the inequality and Established, Represents the expected displacement trajectory variable. The derivative of the expected displacement trajectory variable. ;

[0097] The error constraints of the robot system module include tracking error constraints and virtual error surface constraints:

[0098]

[0099] in, Represents the tracking error constraint conditions, It is the tracking error obtained from real-time measurement. Represents the output and state variables of a first-order filter. deviation, This represents the virtual error surface constraint condition.

[0100] The filter module is established as follows:

[0101] by Based on this, a virtual error surface is established, Mapping to a new coordinate space; introducing a first-order filter to solve for the virtual control signal of the virtual controller fuzzy control;

[0102] The virtual error surface is as follows:

[0103]

[0104]

[0105]

[0106] in, It is the expected trajectory of joint movement. It is the output error of the first-order filter; The tracking error is the first One portion, It is the virtual error surface. One portion, The output error of the first-order filter is the... One component;

[0107] This represents the output of the virtual control signal after passing through a first-order filter.

[0108] The first-order filter is as follows:

[0109]

[0110] in, It is the time constant in the filter module.

[0111] The virtual controller module is established as follows;

[0112] The virtual controller module and its component form are as follows:

[0113]

[0114]

[0115] In the formula, This represents the constraint boundary repulsion term in the virtual controller module. , The tracking error constraint is the first One portion, , ;in, This refers to the feedback gain parameter in the virtual controller module. It is a time-varying gain parameter. make sure Bounded, satisfying the inequality , .

[0116] The fuzzy adaptive law module is obtained based on the fuzzy logic system;

[0117] The fuzzy logic system is specifically as follows: It is a continuous function defined on a closed set. Above, using fuzzy logic systems express Satisfies the expression:

[0118]

[0119] in, Represents the fuzzy fundamental function vector. Indicates the upper limit. It is the optimal approximation error. Represents the ideal weight. Represents the total number of fuzzy rules, the th fuzzy fundamental function vectors Definition:

[0120]

[0121] It is a fuzzy membership function. The input vector representing the fuzzy membership function. The total number of input variables in the fuzzy logic system;

[0122] Based on the approximation properties of fuzzy logic systems, this method handles unknown nonlinear functions. The specific form of the components is as follows:

[0123]

[0124] In the formula, The positive definite symmetric inertial matrix is ​​the first... One portion, For external disturbances One portion, The Coriolis force-centrifugal force matrix is ​​the first... One portion, For the Jacobian matrix, the first... One portion, The gravity vector matrix is ​​the first One portion, For the ideal weights A vector, , The optimal approximation error is represented by the first... Each component is bounded. , This represents the upper bound of the optimal approximation error. This represents the square of the norm of the weight vector. Represents the optimal parameters The estimated value;

[0125] The mathematical model of the fuzzy adaptive law module is as follows:

[0126]

[0127] In the formula, , representing the constraint boundary repulsion term in the fuzzy adaptive law. The virtual error surface constraint condition is the first One portion, , These are the design parameters in the fuzzy adaptive law. These are the adaptive parameters of the fuzzy adaptive law; It is the dynamic change information of the fuzzy adaptive law parameter module, which will The input is fed into the fuzzy adaptive controller module.

[0128] The mathematical model of the fuzzy adaptive controller module is as follows:

[0129]

[0130] No. The components are:

[0131]

[0132] In the formula, Represents virtual error margin and The coupling factor corresponding to the gradient of the barrier potential field. It is the first One coupling factor component, The first part representing the linear part of the dead zone model after inverse transformation One portion, , , The feedback gain parameter in the fuzzy adaptive controller. The first-order filter represents the output of the virtual control signal. One derivative component, It is a time-varying gain parameter. make sure Bounded, satisfying the inequality , ;

[0133] The output signal of the fuzzy adaptive controller module is input to the robot system module to calculate the control torque of the actuator of the robot system module.

[0134] like Figure 1 As shown, the input of the filter module is connected to the output of the robot system module containing dead-zone characteristics and uncertainties; the output of the filter module is connected to the input of the virtual controller module; the output of the virtual controller module is connected to the input of the fuzzy adaptive law module; the output of the fuzzy adaptive law module is connected to both the virtual controller module and the fuzzy adaptive controller module; and the output of the fuzzy adaptive controller module is connected to the input of the robot system module containing dead-zone characteristics and uncertainties.

[0135] The fuzzy adaptive law module is used to calculate the dynamic changes of the adaptive parameters and send the dynamic changes to the adaptive controller module.

[0136] Overall process working principle:

[0137] This invention presents a fuzzy adaptive control method for uncertain robot system constraints, with the structure as follows: Figure 1 As shown. During robot operation, joint displacement... ,speed The information is input to the filter module, and the result is... The input is fed to the virtual controller module for calculation, and the virtual controller obtained by the virtual controller module... The values ​​are fed into the input terminal of the fuzzy control adaptive law module for calculation, resulting in estimated values ​​of the fuzzy adaptive law parameters. The data information is transmitted to the fuzzy adaptive controller module to calculate the control signal. The control signal of the fuzzy adaptive controller is then fed to the system dead zone model to obtain the control torque. Adjust the state information of the robot system model to achieve fuzzy adaptive state feedback control.

[0138] The simulation results are as follows:

[0139] pass Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 As shown, under fuzzy adaptive control, the displacement and velocity of the joints in an uncertain robot system with dead-zone characteristics achieve good tracking performance within the boundary constraints. The adaptive control signal and adaptive law parameters exhibit good control performance, and the tracking error converges. In other words, the proposed control method effectively controls an uncertain robot system with dead-zone characteristics.

[0140] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A fuzzy adaptive control method for uncertain robot system constraints, characterized in that, Includes the following steps: Step 1: Integrate the dead zone model into the uncertain robot dynamics equations to establish a robot system module; construct time-varying constraints and error constraints for the robot system module to limit joint displacement, joint velocity, tracking error, and virtual error surface; the robot system module obtains the output signal of the fuzzy adaptive controller module, calculates the control torque of the robot system module's actuator, and adjusts the robot's joint displacement according to the control torque; Step 2: The filter module obtains the robot's joint displacement, velocity, and acceleration output by the robot system module, and calculates the output signal of the first-order filter constructed based on the virtual error surface; the virtual controller module obtains the output signal of the first-order filter of the filter module, and calculates the virtual control signal for fuzzy control. Step 3: The fuzzy adaptive law module acquires the virtual control signal of the fuzzy control and calculates the dynamic change information of the fuzzy adaptive law module; Step 4: The fuzzy adaptive controller module acquires dynamic change information and calculates the output signal of the fuzzy adaptive controller module.

2. The fuzzy adaptive control method for uncertain robot system constraints according to claim 1, characterized in that, The robot dynamics equations for the uncertainty mentioned in step 1 are as follows: in, , , These represent joint displacement, joint velocity, and joint acceleration, respectively. Represents a positive definite symmetric inertial matrix. Represents the unknown Coriolis force-centrifugal force matrix. The gravity vector matrix, It is an unknown Jacobian matrix, subject to external disturbances. , It is the upper bound of the disturbance. It is a control signal; The dead zone model is as follows: in, Represents the input to the control law. The right slope represents the dead zone feature. The left slope represents the dead zone feature. Represents the right intercept point. Represents the left intercept point. It is the actuator control torque at time t; A smooth and continuous mathematical model is used to perform an inverse transformation of the feedback linearization dead zone, and the dead zone model is transformed into the following form: In the formula, , in, This represents the linear part of the dead zone model after inverse transformation. This represents the nonlinear part of the dead-zone model after inverse transformation. Bounded, Represents the upper bound of the cutoff point; Define state variables , Substituting the inversely transformed dead-zone model into the uncertain robot dynamics equations, we obtain the following expression for the robot system module: in, , This represents the displacement of the i-th joint, where i is from 1 to n; , Represents the velocity of the i-th joint; It is the control torque of the robot system module actuator; Represents external disturbances related to time; Represents the state variable The gravity vector matrix, It is the output feedback variable; The time-varying constraint condition is: , ;in, and It is a continuous, differentiable, time-varying constraint bound, designed based on the physical limitations of the robot, and used to characterize safety requirements that change over time; The robot system module assumes the existence of constants. and Satisfying the inequality and , ; and The i-th joint represents a continuous, differentiable, time-varying constraint bound; the robot system module introduces a function vector. and And satisfy the inequality and And the function components and , make the inequality and Established, Represents the expected displacement trajectory variable. The derivative of the expected displacement trajectory variable. ; The error constraints of the robot system module include tracking error constraints and virtual error surface constraints: ， ， in, This represents the tracking error constraint. It is the tracking error obtained from real-time measurement. Represents the output and state variables of a first-order filter. deviation, This represents the virtual error surface constraint condition.

3. The fuzzy adaptive control method for uncertain robot system constraints according to claim 2, characterized in that, The filter module is established as follows: by Based on this, a virtual error surface is established, Mapping to a new coordinate space; introducing a first-order filter to solve for the virtual control signal of fuzzy control; The virtual error surface is as follows: in, It is the expected trajectory of joint movement. It is the output error of the first-order filter; The tracking error is the first One portion, It is the virtual error surface. One portion, The output error of the first-order filter is the... One component; This represents the output of the virtual control signal after passing through a first-order filter. Virtual control signals representing fuzzy control; The first-order filter is as follows: in, It is the time constant in the filter module.

4. The fuzzy adaptive control method for uncertain robot system constraints according to claim 2, characterized in that, The virtual controller module is established as follows; The virtual controller module and its component form are as follows: In the formula, This represents the constraint boundary repulsion term in the virtual controller module. , The tracking error constraint is the first One portion, The i-th component of the virtual control signal represents the fuzzy control signal. , ;in, This refers to the feedback gain parameter in the virtual controller module. It is a time-varying gain parameter. make sure Bounded, satisfying the inequality , .

5. The fuzzy adaptive control method for uncertain robot system constraints according to claim 4, characterized in that, The fuzzy adaptive law module is obtained based on the fuzzy logic system; The fuzzy logic system is specifically as follows: It is a continuous function defined on a closed set. Above, using fuzzy logic systems express Satisfies the expression: in, Represents the fuzzy fundamental function vector. Indicates the upper limit. It is the optimal approximation error. Represents the ideal weight. Represents the total number of fuzzy rules, the th fuzzy fundamental function vectors Definition: It is a fuzzy membership function. The input vector representing the fuzzy membership function. The total number of input variables in the fuzzy logic system; Based on the approximation properties of fuzzy logic systems, this method handles unknown nonlinear functions. The specific form of the components is as follows: In the formula, The positive definite symmetric inertial matrix is ​​the first... One portion, For external disturbances One portion, The Coriolis force-centrifugal force matrix is ​​the first... One portion, For the Jacobian matrix, the first... One portion, The gravity vector matrix is ​​the first One portion, For the ideal weights A vector, , The optimal approximation error is represented by the first... Each component is bounded. , This represents the upper bound of the optimal approximation error. This represents the square of the norm of the weight vector. Represents the optimal parameters The estimated value; The mathematical model of the fuzzy adaptive law module is as follows: In the formula, , representing the constraint boundary repulsion term in the fuzzy adaptive law. The virtual error surface constraint condition is the first One portion, , These are the design parameters in the fuzzy adaptive law. These are the adaptive parameters of the fuzzy adaptive law; It is the dynamic change information of the fuzzy adaptive law parameter module, which will The input is fed into the fuzzy adaptive controller module.

6. The fuzzy adaptive control method for uncertain robot system constraints according to claim 4, characterized in that, The mathematical model of the fuzzy adaptive controller module is as follows: No. The components are: In the formula, Represents virtual error margin and The coupling factor corresponding to the gradient of the barrier potential field. It is the first One coupling factor component, The first part representing the linear part of the dead zone model after inverse transformation One portion, , , The feedback gain parameter in the fuzzy adaptive controller. , representing the output of the virtual control signal after passing through the first-order filter. One derivative component, It is a time-varying gain parameter. make sure Bounded, satisfying the inequality , ; The output signal of the fuzzy adaptive controller module is input to the robot system module to calculate the control torque of the actuator of the robot system module.

7. A fuzzy adaptive control system for uncertain constraints of a robot system, characterized in that, The fuzzy adaptive control method for uncertain robot system constraints as described in any one of claims 1-6 includes: Robot system modules containing dead zone characteristics and uncertainties, filter modules, virtual controller modules, fuzzy adaptive law modules, and fuzzy adaptive controller modules; The input of the filter module is connected to the output of the robot system module containing dead-zone characteristics and uncertainties; the output of the filter module is connected to the input of the virtual controller module; the output of the virtual controller module is connected to the input of the fuzzy adaptive law module; the output of the fuzzy adaptive law module is connected to the inputs of the virtual controller module and the fuzzy adaptive controller module respectively; the output of the fuzzy adaptive controller module is connected to the input of the robot system module containing dead-zone characteristics and uncertainties. The fuzzy adaptive law module is used to calculate the dynamic change information of the adaptive law parameters and send the dynamic change information to the fuzzy adaptive controller module.

Images