An adaptive robust feedback motion control method for motor-driven machines

Abstract

The application discloses a kind of adaptive robust feedback motion control methods of motor-driven machinery, comprising: setting the desired motion trajectory of control target;System unknown constant estimation value output by adaptive module is obtained, and feedforward compensation module is used to obtain feedforward compensation control input;Robust feedback control input is fed back by robust feedback module, and system control input is obtained;System is run, and the actual motion trajectory of control target is obtained, and system control error is obtained;System control error is used as input: current robust feedback control input is obtained using robust feedback module;Current system unknown constant estimation value is obtained using adaptive module, and the adaptive robust feedback motion control is realized.The application uses adaptive module to continuously adjust feedforward compensation control parameter, so that it accurately describes the characteristics of different working conditions, reduces system tracking error;Robust feedback module is used to obtain feedback control input, suppress external interference, and ensure the transient response performance of system.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology, specifically to an adaptive robust feedback motion control method for motor-driven machinery. Background Technology

[0002] Compared to hydraulic or pneumatic drives, electrically driven machinery offers advantages such as faster motion response, more stable control, and smaller size. It is widely used in pillar industries like medical devices, industrial production, and transportation, as well as strategic sectors like national defense, aerospace, and military. These electrically driven machines typically consist of a motion actuator, a motor, a transmission system, and a control system. The motion actuator is linked to the motor's output shaft. The motor is generally a servo motor, and transmission can be achieved via a lead screw and nut, or a pulley drive. In this case, the motion actuator is fixed to the nut, and motion is achieved by controlling the motor's rotation. Adaptive control is a method of adjusting control parameters based on system feedback information to adapt to dynamic changes and parameter uncertainties in the system. In electrically driven machinery, adaptive control can adjust control parameters based on varying load resistance under different working conditions to achieve consistent motion trajectory and thrust. Robust control is a control method designed to address system parameter uncertainties and external disturbances, ensuring system stability and performance. In electrically driven machinery, robust control can address the impact of external disturbances (such as random environmental vibrations) and system parameter uncertainties (such as mechanical friction and load changes) on motion tracking accuracy. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a motor-driven mechanical motion control method that can be used under different motion load conditions, adaptively adjust feedback control parameters to achieve a set motion trajectory, and resist external interference, and is suitable for complex application scenarios.

[0004] Adaptive robust feedback motion control is designed to ensure that motor-driven machinery can adapt to different motion load conditions and environmental conditions, providing stable and accurate motion. This method combines the concepts of adaptive control and robust control.

[0005] The technical solution adopted by this invention to solve the above-mentioned technical problems is as follows: an adaptive robust feedback motion control method for motor-driven machinery, mainly executed by a feedforward compensation module, an adaptive module, and a robust feedback module, specifically including:

[0006] (1) Set the desired motion trajectory of the control target;

[0007] (2) Obtain the estimated values ​​of the unknown system constants output by the adaptive module, and use the feedforward compensation module to obtain the feedforward compensation control input;

[0008] (3) The system is operated by obtaining the system control input, the actual motion trajectory of the control target is obtained, and the system control error is obtained according to the expected motion trajectory of the control target;

[0009] (4) Using system control error as input: Use the robust feedback module to obtain the current robust feedback control input; use the adaptive module to obtain the current unknown constant estimate of the system, return to step (2), and realize the adaptive robust feedback motion control of the motor-driven machinery.

[0010] During modeling, a feedforward compensation module is first established based on prior knowledge of the model; then an adaptive module is established to adjust the feedforward compensation control parameters in real time; at the same time, a robust feedback module is built to adjust the feedback control input according to the system error.

[0011] In this invention, the control target is the motion actuator of the motor-driven machinery.

[0012] The motion actuator refers to the mechanism in a mechanical structure that is responsible for ultimately realizing the motion to complete the work task. Common motor-driven machines include chest compression machines for cardiopulmonary resuscitation, mold opening and closing mechanisms and injection mechanisms in injection molding machines in the plastics industry, and their corresponding actuators are chest compression heads, moving templates and injection screws, respectively.

[0013] As a further preferred embodiment, the motor-driven mechanism is a chest compression machine, and the control target is the compression head of the chest compression machine.

[0014] Before constructing the feedforward compensation module, it is necessary to select a corresponding model structure based on the structural characteristics of the system. Preferably, this invention selects a semi-strict feedback model as the system model structure. Preferably, the feedforward compensation module is established based on the system dynamics model, where a semi-strict feedback model is selected.

[0015] Preferably, the semi-strict feedback model structure is as follows:

[0016] (1)

[0017] Where i = 1, 2, 3…n-1, and n is the total number of state variables in the corresponding system. For the first One system state variable, for Differentiating with respect to time, For the unknown constants related to the i-th state variable, Let be the shape function of the known model features related to the 1st to 1st state variables in the i-th state variable at time t. For time t The nonlinear characteristics of systems that cannot be modeled in Chinese Let x be at time t n In, with x1~x n The nonlinear characteristics of the system that cannot be modeled. To control the impact of input on the system, For system control input, For the unknown constants related to the nth state variable, Let y be the shape function of the known model features related to the 1st to nth state variables in the nth state variable at time t, and let y be the system output. The trajectory information corresponding to the controlled target.

[0018] Preferably, the feedforward compensation module has the following structure:

[0019]

[0020] This is the input for feedforward compensation control. To control the velocity of the target at each moment in the desired trajectory, Given the shape functions of the system model with known characteristics, These are the estimated values ​​of the unknown constants in the model after mapping.

[0021] Preferably, the feedforward compensation module can use a semi-strict feedback model to describe the model characteristics. Here, a sufficiently smooth shape function represents the known system model characteristics obtained from prior knowledge, an unknown constant matrix represents the unknown linear characteristics of the system, and unknown nonlinear equations represent all system nonlinearities that cannot be modeled.

[0022] Preferably, the system control input is the sum of the feedforward compensation control input and the robust feedback control input.

[0023] Preferably, for more complex systems, a recursive control law can be used in the adaptive module to establish the adaptive variation law of the feedforward compensation model. Preferably, the control parameters obtained through adaptive feedback can be constrained within their range through smooth projection to prevent system divergence.

[0024] Furthermore, in the adaptive module, the parameter adaptive law is used to update the unknown constants of the system. The updated values ​​of the unknown constants of the system are constrained by smooth projection, and finally the estimated values ​​of the unknown constants of the system of the feedforward compensation module are obtained.

[0025] As a further optimization, the parameter adaptation law in the adaptive module is as follows:

[0026] Adaptive law ;

[0027] Adaptive law ;

[0028] Adaptive law ;

[0029] in, , , A constant greater than 0 , , Given the shape functions of the system model with known characteristics, , , This is for system control error;

[0030] Smooth projection uses the following mapping relationship:

[0031]

[0032] in, It is a constant diagonal matrix. for The adaptive law, for Differentiating with respect to time, These are the estimated values ​​of the unknown constants in the model after mapping.

[0033] In the adaptive and feedforward compensation modules, recursive control laws are used to establish the corresponding adaptive control laws; in the robust feedback module, recursive control laws are used to establish the robust control laws. As a further preferred option, the robust feedback module can use a combination of linear stable feedback and nonlinear robust feedback to determine the feedback control quantity.

[0034] Preferably, the adaptive control law is:

[0035]

[0036] in, The transpose of the estimate of the unknown constants associated with the nth state variable. Let x1~x2 be the nth state variable at time t and the nth state variables x1~x2~x3~x4~x5~x6~x7~x8~x9~x1 ...1~x1~x1~ n Shape functions of the known model features. for The computable part in The pair obtained in the recursive control law The control law, for Differentiate with respect to time;

[0037] The recursive control law is as follows:

[0038]

[0039]

[0040] in, The pair obtained in the recursive control law The control law, For the first state variable The transpose of the estimates of the relevant unknown constants, This generally corresponds to the motion trajectory information of the controlled target (for example, a chest compression machine). This generally corresponds to the motion trajectory information of the pressing head. Let x1 be the shape function of the known model feature related to the first state variable x1 in the first state variable at time t. To control the velocity of the target at each moment in the desired trajectory, For the first state variable The minimum value of the relevant unknown constants, It is a constant. It is a constant diagonal matrix. It is a positive definite constant diagonal matrix. It is a positive integer. With the first state variable The maximum value of the relevant unknown constants, This represents the maximum value of the nonlinear environmental disturbance, i.e. , For the first state variable That is, the error between the actuator's motion trajectory and the desired motion trajectory; The pair obtained in the recursive control law The control law, For the i-th state variable The transpose of the estimates of the relevant unknown constants, Let x be the state variable at time t and the state variable at time t. i Shape functions of the known model features. The pair obtained in the recursive control law The computable part of the control law For the i-th state variable The minimum value of the relevant unknown constants, It is a constant. It is a constant diagonal matrix. It is a positive definite constant diagonal matrix. It is a positive integer. With the i-th state variable The maximum value of the relevant unknown constants, For the i-th state variable The error between the expected trajectory and the actual trajectory.

[0041] The robust control law is as follows:

[0042]

[0043]

[0044]

[0045] in: This refers to the control input for the linear feedback section. This refers to the control input for the nonlinear robust feedback component. .

[0046] Let be the minimum value of the unknown constants related to the nth system state variable. For positive integers, and The conditions that need to be met are shown in the following formula:

[0047]

[0048] The control method of this invention can control the motion trajectory of the actuator of the motor-driven machinery, including the motion distance and motion frequency; it uses an adaptive module to continuously adjust the feedforward compensation control parameters to accurately describe the characteristics of different working conditions and reduce the system tracking error; and it uses a robust feedback module to obtain feedback control input, suppress external interference, and ensure the transient response performance of the system.

[0049] In summary, the adaptive robust feedback motion control method for motor-driven machinery of the present invention combines adaptive control and robust control techniques. By adaptively adjusting control parameters and suppressing external disturbances, it provides stable and accurate motion, thereby improving the effect of motion control. Attached Figure Description

[0050] Figure 1 This is a flowchart of the adaptive robust feedback motion control method for motor-driven machinery according to the present invention.

[0051] in: Expected trajectory; Expected speed of movement; The characteristics of a known system model expressed using fully smooth shape functions; The unknown linear characteristics of the system are represented by an unknown constant matrix; , the predicted value of the unknown linear characteristics of the system; for Differentiating with respect to time, The estimated values ​​of the unknown constants of the model after smooth projection; All nonlinear characteristics of systems that cannot be modeled; This indicates the effect of control inputs on the system; Control input; Feedforward compensation control input; Robust feedback control input; The control input of the linear stability feedback section; The control input of the nonlinear robust feedback section; System output; System control error; Linear stable feedback gain; Adaptive law for controlling parameters; , The adaptive law; , smooth projection function.

[0052] Figure 2 A schematic diagram of a typical transmission structure for motor-driven machinery.

[0053] Figure 3 Parameter estimates obtained by different control methods under undisturbed conditions.

[0054] Figure 4 Motion control errors under different control methods in the absence of disturbance.

[0055] Figure 5 Parameter estimates obtained by different control methods under disturbance conditions.

[0056] Figure 6 Motion control errors under different control methods in the presence of disturbances. Detailed Implementation

[0057] like Figure 2 As shown, a typical motor-driven mechanical system mainly includes an actuator that performs the motion, a drive motor that drives the actuator, a control system (such as a computer or integrated chip), and some supporting transmission components (including pulleys, lead screws, nuts, etc.). The motion trajectory of the actuator is the final control target, ultimately achieving precise motion. For example, in a common chest compression system, the actuator includes a compression head, a drive motor that drives the compression head, a control system (such as a computer or integrated chip), and some supporting transmission components (including pulleys, lead screws, nuts, etc.). This invention uses the control system of this structure as an example for illustration, but this invention is also applicable to motor-driven mechanical systems of other structures. A detailed flowchart of the method is shown below. Figure 1As shown, it is mainly executed by the feedforward compensation module, the adaptive module, and the robust feedback module. The specific implementation includes the following steps:

[0058] (1) Set the desired motion trajectory of the control target;

[0059] (2) Obtain the estimated values ​​of the unknown system constants output by the adaptive module, and use the feedforward compensation module to obtain the feedforward compensation control input;

[0060] (3) The system is operated by obtaining the system control input, the actual motion trajectory of the control target is obtained, and the system control error is obtained according to the expected motion trajectory of the control target;

[0061] (4) Using system control error as input: Use the robust feedback module to obtain the current robust feedback control input; use the adaptive module to obtain the current unknown constant estimate of the system, return to step (2), and realize the adaptive robust feedback motion control of the motor-driven machinery.

[0062] Furthermore, taking the chest compression system as an example, the specific aspects of model selection, construction, and data processing include the following:

[0063] Step 1: Establish a model of the controlled system and obtain the initial parameters of the model. The specific process is as follows:

[0064] Step 1-1: Establish a semi-strict feedback model of the controlled motor-driven mechanical system, as shown in equation (1):

[0065] (1)

[0066] Where i = 1, 2, 3…n-1, and n is the total number of state variables in the corresponding system. For the first One system state variable, For the first One system state variable, for Differentiating with respect to time, for Differentiating with respect to time, This generally corresponds to the motion trajectory information of the actuator (i.e., the pressing head); These are unknown constants in the system model. , For the unknown constants related to the i-th state variable, Given the shape functions of the system model with known characteristics, Let t be the shape function of the known model features related to the 1st to 1st state variables in the ith state variable. Let x be at time t i Nonlinear characteristics of systems that cannot be modeled in Chinese To control the impact of input on the system, To control the impact of input on the system, To control the input, For the unknown constants related to the nth state variable, Let n be the state variable at time t. In the diagram, the first to nth state variables x1~x n Shape functions of the known model features. At time t, x n In and x1~x n The nonlinear characteristics of the system that cannot be modeled. This is the system output.

[0067] For practical chest compression machines, the system state variables involved mainly include motor rotation angle, motor torque, pulley rotation angle, lead screw rotation angle, compression head position, compression head resistance, etc. The control input is generally the motor current, and the system output is generally the movement trajectory of the compression head.

[0068] In the model structure, These are unknown constants in the system model and are an important component of the system model's characteristics. However, due to the high complexity of the system or the difficulty in specific measurement, they cannot be directly obtained in practical applications. In chest compressions, they specifically manifest as the resistance to chest compressions, the actual transmission ratio and efficiency between various transmission links, friction, vibration, etc.

[0069] The introduction of this method enables a more accurate description of the characteristics of different pressing objects, reducing system tracking errors.

[0070] Step 1-2: Set the range of values ​​for the unknown constants in the system model, as shown in equation (2).

[0071] (2)

[0072] in, for The minimum value, for The maximum value is obtained by using this step to obtain all θ( The corresponding maximum and minimum values.

[0073] Step 2: Set the desired movement trajectory of the pressing head. This generally includes the depth and frequency of the pressing head. The speed of movement corresponding to the pressing head, i.e. The derivative with respect to time.

[0074] Step 3: Based on Step 1-1, select the model structure and establish the feedforward compensation module. The specific process is as follows:

[0075] Step 3-1: Calculate the velocity at each moment in the desired trajectory. .

[0076] Step 3-2: Set unknown constants The estimated initial value is shown in equation (3):

[0077] (3)

[0078] Step 3-3: Establish the feedforward compensation module based on the system dynamics model in Step 1-1, as shown in Equation (4):

[0079] (4)

[0080] in, This is the feedforward compensation control quantity (i.e., the feedforward compensation control input). Let the velocity be the velocity at each moment in the desired trajectory. Given the shape functions of the system model with known characteristics, These are the estimated values ​​of the unknown constants in the model after mapping; this mapping relationship will be established in the adaptive module. This means calculating the rate of change of the system output based on the system model.

[0081] Step 4: Establish the recursive adaptive law and derive the control parameters in the adaptive module and the robust feedback module.

[0082] Step 4-1: Establish the mapping relationship for parameter estimation.

[0083] In the adaptive module, the parameter adaptive law is used to obtain... The updated values ​​are then smoothed to obtain estimates of the unknown constants of the current model.

[0084] make Indicates the unknown parameter The estimated value, The error in parameter estimation can be represented by the mapping shown in equation (5):

[0085] (5)

[0086] in, This represents the mapping relationship for parameter estimation.

[0087] set up for The adaptive law, therefore for The parameter estimate can be expressed by equation (6):

[0088] (6)

[0089] in, It is a constant diagonal matrix. for The adaptive law will be established later.

[0090] The mapping relationship in equation (6) can guarantee that:

[0091] (7)

[0092] This indicates that for any τ.

[0093] For the prediction of unknown constants of the system The adaptive law is an algorithm that adaptively adjusts and calculates the predicted values ​​of unknown system constants based on the system control error. When the system control error is large, it indicates that the current estimation of the unknown system constants is inaccurate. Therefore, it is necessary to continuously adjust the predicted values ​​using the adaptive law until the control error is zero.

[0094] Step 4-2, based on the system model Formula, assumption For its virtual input, derive the derivation for virtual input control law .

[0095] In the system model The formula is shown in formula (8):

[0096] (8)

[0097] by If the virtual control target is the same as the actual system control target, then its control error is as shown in equation (9):

[0098] (9)

[0099] in, For virtual control target The control error, for The expected value.

[0100] According to equation (8) The rate of change can be expressed by the formula:

[0101] (10)

[0102] Assumption Its virtual input, the control target is to make If the control error is minimized, then for virtual input control law It consists of two parts, as shown in equation (11):

[0103] (11)

[0104] in, For adaptive control law, This is a robust control law.

[0105] The function of the adaptive control law is to ensure steady-state response performance, according to equation (10) (let... =0), ignoring the unknown nonlinear characteristics that cannot be calculated in the formula ( If =0), then the adaptive control law It can be written as:

[0106] (12)

[0107] Adaptive law It can be calculated from (13):

[0108] (13)

[0109] in, It is a constant.

[0110] The robust control law consists of two parts, as shown in equation (14):

[0111] (14)

[0112] in For the linear feedback part, This is the robust feedback section.

[0113] Linear feedback section It can be expressed by equation (15):

[0114] (15)

[0115] in, It can be expressed by equation (16):

[0116] (16)

[0117] in, It is a constant. It must be a positive definite constant diagonal matrix, which needs to satisfy certain conditions, which will be introduced later. for The minimum value. This indicates the search for the 2-norm.

[0118] Robust Feedback Section The following conditions must be met, as shown in equation (17):

[0119] (17)

[0120] Nonlinearity of all unmodelable systems expressed by unknown nonlinear equations. It is believed that setting a normal number can suppress the influence of external interference to a small range;

[0121] Therefore, As shown in equation (18):

[0122] (18)

[0123] The maximum value of the nonlinear disturbance experienced by the system is, i.e. .

[0124] therefore, It can be expressed by equation (19):

[0125] (19)

[0126] Step 4-3: Based on the system model Formula, assumption For its virtual input, derive the derivation for virtual input control law , .

[0127] In the system model The formula is shown in formula (20):

[0128] (20)

[0129] by If the virtual control target is the same as the actual system control target, then its control error is as shown in equation (21):

[0130] (twenty one)

[0131] in, For virtual control target The control error, The pair obtained in the recursive control law The control law.

[0132] According to equation (20). The rate of change can be expressed by the formula:

[0133] (twenty two)

[0134] in, for The computable part; for The incalculable part mainly consists of terms containing parameter estimation errors, unknown nonlinear characteristics, and the derivative of the parameter estimation, as shown in equation (23).

[0135] (twenty three)

[0136] in, for The computable part in the equation, that is, the part that does not contain unknown nonlinear characteristics, is shown in equation (24):

[0137] (twenty four)

[0138] Assumption Its virtual input, the control target is to make To minimize the control error, a control law needs to be designed to handle the nonlinear quantities in equation (22), including... and For virtual input control law It also contains two parts, as shown in equation (25):

[0139] (25)

[0140] in, For adaptive control law, The robust control law is shown in equation (26): (26)

[0142] in,

[0143] (27)

[0144] in, For positive integers, and It is a positive definite diagonal matrix.

[0145] Adaptive law It can be calculated using equation (28):

[0146] (28)

[0147] in, .

[0148] Robust Feedback Section As shown in equation (29):

[0149] (29)

[0150] in, It is a normal number set by humans.

[0151] Step 4-4: Through the above steps, the control input is derived step by step. The control law is shown in equation (30):

[0152] (30)

[0153] in,

[0154] for The computable part; The pair obtained in the recursive control law The control law;

[0155] (31)

[0156] For positive integers, and It is a positive definite diagonal matrix;

[0157] Adaptive law for:

[0158] (32)

[0159] in, .

[0160] Positive definite diagonal matrix and The conditions that need to be met are shown in equation (33):

[0161] (33)

[0162] Step 5: Establish an adaptive module to calculate the estimated values ​​of unknown parameters based on system errors.

[0163] Based on the adaptive law derived above, substituting it into equation (6) yields the estimated value of the unknown parameter as shown in equation (34):

[0164] (34)

[0165] in, , .

[0166] Establish adaptive control laws for the adaptive module and the feedforward compensation module:

[0167]

[0168] Step 6: Establish a robust feedback module and calculate the feedback control quantity based on the system error.

[0169] Step 6-1: Establish the linear feedback component.

[0170] According to equation (30), the linear feedback part can be obtained as follows:

[0171] (35)

[0172] in, This is the control variable for the linear feedback section.

[0173] Step 6-2: Establish the nonlinear robust feedback part.

[0174] According to equation (30), the nonlinear robust feedback part can be obtained as follows:

[0175] (36)

[0176] in, This represents the control input for the nonlinear robust feedback component.

[0177] Simulation Experiment

[0178] The method proposed in this invention was verified through simulation experiments. In the simulation experiments, the transmission structure of the press machine was simplified, assuming that the simulation system contains only one state variable, namely the motion trajectory of the pressing head, and the shape functions of the system model are known. Assuming it is a sine function, the unknown constants in the system model are... Assuming it is the amplitude coefficient of a sine function. The control object model used is shown in equation (37):

[0179] (37)

[0180] in, These are unknown constants in the system model. To simplify the movement speed of the pressing head in the system, To simplify the movement trajectory of the pressing head in the system, This refers to nonlinear features in the system that cannot be modeled.

[0181] (38)

[0182] The desired trajectory is shown in equation (39):

[0183] (39)

[0184] Simulation experiments were conducted using adaptive control (containing only the adaptive module of this invention, referred to as AdaptiveControl in the figure), robust control (containing only the robust module of this invention, referred to as Robust Control), and the adaptive robust control method proposed in this invention (referred to as Adaptive Robust Control in the figure). The control parameters were set as shown in the equation:

[0185] (40)

[0186] In the absence of nonlinear perturbations, i.e. Different methods for unknown parameters The estimated value is as follows Figure 3 As shown, the control errors under different methods are as follows: Figure 4 As shown in the figure, the adaptive robust control method exhibits the fastest parameter estimation convergence, zero steady-state error, and fastest transient response, demonstrating the best control performance in the absence of external disturbances.

[0187] In the presence of external nonlinear interference, set The parameter estimates obtained under different control methods are as follows: Figure 5 As shown, the motion control errors under different control methods are as follows: Figure 6 As shown in the figure, under external nonlinear disturbances, the parameter estimation under adaptive robust control exhibits the smallest fluctuation, stabilizing near the actual parameter values. Furthermore, the absolute value of the motion control error is minimized, and its fluctuation is also minimal. Therefore, adaptive robust control can effectively handle the position parameters in the system and effectively suppress external nonlinear disturbances, ensuring good control performance.

Claims

1. An adaptive robust feedback motion control method for motor-driven machinery, characterized in that, It is mainly executed by the feedforward compensation module, the adaptive module, and the robust feedback module, specifically including: (1) Set the desired motion trajectory of the control target; (2) Obtain the estimated values ​​of the unknown system constants output by the adaptive module, and use the feedforward compensation module to obtain the feedforward compensation control input; (3) The system is operated by obtaining the system control input, the actual motion trajectory of the control target is obtained, and the system control error is obtained according to the expected motion trajectory of the control target; (4) Using system control error as input: Use the robust feedback module to obtain the current robust feedback control input; use the adaptive module to obtain the current unknown constant estimate of the system, return to step (2), and realize the adaptive robust feedback motion control of the motor-driven machinery; The system control input is the sum of the feedforward compensation control input and the robust feedback control input; In the adaptive module and the feedforward compensation module, the corresponding adaptive control law is established using a recursive control law; in the robust feedback module, the robust control law in the robust feedback module is established using a recursive control law. The adaptive control law is as follows: ； in, The transpose of the estimate of the unknown constants associated with the nth state variable. Let x1~x2 be the nth state variable at time t and the nth state variables x1~x2~x3~x4~x5~x6~x7~x8~x9~x1 ... n Shape functions of the known model features. for The computable part in The pair obtained in the recursive control law The control law, for Differentiate with respect to time; The recursive control law is as follows: ； ； in, The pair obtained in the recursive control law The control law, For the first state variable The transpose of the estimates of the relevant unknown constants, Corresponding to the motion trajectory information of the controlled target, Let x1 be the shape function of the known model feature related to the first state variable x1 in the first state variable at time t. To control the velocity of the target at each moment in the desired trajectory, For the first state variable The minimum value of the relevant unknown constants, It is a constant. It is a constant diagonal matrix. It is a positive definite constant diagonal matrix. It is a positive integer. For the first state variable The maximum value of the relevant unknown constants, This represents the maximum value of the nonlinear environmental disturbance, i.e. , For the first state variable That is, to control the error between the target's trajectory and the desired trajectory; The pair obtained in the recursive control law The control law, For the i-th state variable The transpose of the estimates of the relevant unknown constants, Let x be the state variable at time t and the state variable at time t. i Shape functions of the known model features. The pair obtained in the recursive control law The computable part of the control law For the i-th state variable The minimum value of the relevant unknown constants, It is a constant. It is a constant diagonal matrix. and It is a positive definite constant diagonal matrix. It is a positive integer. For the i-th state variable The maximum value of the relevant unknown constants, For the i-th state variable The error between its expected trajectory and the actual trajectory; The robust control law is: ； ； ； in: This refers to the control input for the linear feedback section. This refers to the control input for the nonlinear robust feedback component. ； Let be the minimum value of the unknown constants related to the nth system state variable. Let n be the maximum value of the unknown constants related to the nth system state variable. For positive integers, and The conditions that need to be met are shown in the following formula: 。 2. The adaptive robust feedback motion control method for motor-driven machinery according to claim 1, characterized in that, The control target is the motion actuator of the motor-driven machinery.

3. The adaptive robust feedback motion control method for motor-driven machinery according to claim 1, characterized in that, The feedforward compensation module is established based on the system dynamics model, which is a semi-strict feedback model.

4. The adaptive robust feedback motion control method for motor-driven machinery according to claim 3, characterized in that, The structure of the semi-strict feedback model is as follows: （1） Where i = 1, 2, 3…n-1, and n is the total number of state variables in the corresponding system. For the first One system state variable, for Differentiating with respect to time, For the unknown constants related to the i-th state variable, Let be the shape function of the known model features related to the 1st to 1st state variables in the i-th state variable at time t. At time t The nonlinear characteristics of systems that cannot be modeled in Chinese Let x be at time t n In, with x1~x n The nonlinear characteristics of the system that cannot be modeled. To control the impact of input on the system, For system control input, For the unknown constants related to the nth state variable, Let y be the shape function of the known model features related to the 1st to nth state variables in the nth state variable at time t, and let y be the system output. The trajectory information corresponding to the controlled target.

5. The adaptive robust feedback motion control method for motor-driven machinery according to claim 4, characterized in that, The structure of the feedforward compensation module is as follows: ； This is the input for feedforward compensation control. To control the velocity of the target at each moment in the desired trajectory, Given the shape functions of the system model with known characteristics, These are the estimated values ​​of the unknown constants in the model after mapping.

6. The adaptive robust feedback motion control method for motor-driven machinery according to claim 1, characterized in that, In the adaptive module, the unknown constants of the system are estimated using the parameter adaptive law. The estimated values ​​of the unknown constants of the system are constrained by smooth projection, and finally the estimated values ​​of the unknown constants of the system in the feedforward compensation module are obtained.

7. The adaptive robust feedback motion control method for motor-driven machinery according to claim 6, characterized in that, In the adaptive module, the parameter adaptation law is as follows: Adaptive law ; Adaptive law ; Adaptive law ; in, , , A constant greater than 0 , , Given the shape functions of the system model with known characteristics, , , This is for system control error; Smooth projection uses the following mapping relationship: ； in, It is a constant diagonal matrix. for The adaptive law, for Differentiating with respect to time, These are the estimated values ​​of the unknown constants in the model after mapping.

Images