Method and system for scheduled time tracking control of morphing aircraft with unknown control coefficients

Abstract

The application provides a predetermined time tracking control method and system for a morphing aircraft with unknown control coefficients, and belongs to the technical field of aircraft attitude control. The method comprises the following steps: a tracking error system containing unknown time-varying control coefficients is constructed, a predetermined time performance function and an error transformation mechanism are introduced, and a constrained tracking error is converted into an unconstrained variable; an enhanced variable-frequency Nussbaum function and a state updating mechanism are designed to adaptively compensate for unknown control directions; a dynamic surface backstepping structure of a command filter is used to avoid repeated derivation, and a switching neural network is combined to online approximate unknown nonlinear dynamics; and actual control inputs are synthesized. The predetermined time tracking control method and system for the morphing aircraft with unknown control coefficients can guarantee that tracking errors converge within a user-preset time and remain within a given performance boundary, and effectively suppress model uncertainty and external disturbances under the condition that the sign of the control coefficient is unknown and rapidly time-varying.

Description

Technical Field

[0001] This invention relates to the field of aircraft attitude control technology, and in particular to a method and system for time-tracking control of deformable aircraft with unknown control coefficients. Background Technology

[0002] Deformable aircraft can achieve multi-mission adaptability and better aerodynamic performance by changing their configuration during flight. However, configuration deformation causes significant changes in aerodynamic derivatives, aerodynamic center position and moment distribution over time, making the longitudinal attitude channel exhibit strong nonlinearity, time-varying, coupling and multi-source uncertainty. At the same time, changes in the flight envelope and external disturbances further increase the difficulty of control.

[0003] Existing adaptive control methods for deformable aircraft often rely on prior knowledge of the control direction (i.e., the sign of the control gain) or require that the unknown control directions of multiple control channels satisfy the consistency assumption. The applicability of these methods is limited when multiple unknown time-varying control coefficients exist and their control directions may be inconsistent. Furthermore, common finite-time / fixed-time control methods struggle to directly specify the convergence time when the error enters the preset performance region; rigorous feedback system designs based on backstepping are prone to "derivative explosion" in high-order scenarios and have limited ability to compensate for unmodeled dynamics and time-varying uncertainties.

[0004] Therefore, there is an urgent need to develop a deformable aircraft attitude tracking control method that can simultaneously address unknown control direction, preset convergence time, avoid computational explosion, and effectively compensate for strong time-varying nonlinear uncertainties, so as to improve its autonomous performance and mission reliability in complex and dynamic environments. Summary of the Invention

[0005] The purpose of this invention is to provide a predetermined time tracking control method and system for deformable aircraft with unknown control coefficients, which solves key technical problems such as unknown control direction, unpredictable convergence time, high computational complexity, and insufficient compensation for strong time-varying nonlinear uncertainties.

[0006] To achieve the above objectives, the present invention provides a predetermined time tracking control method for a deformable aircraft with unknown control coefficients, comprising the following steps: Step S1: Obtain the measurable state variables and reference command trajectory of the longitudinal attitude channel of the deformable aircraft. Based on the time-varying characteristics of aerodynamic parameters, the longitudinal dynamics are abstracted into a time-varying nonlinear system with unknown time-varying control coefficients, and the system tracking error is defined. Step S2: Introduce a predetermined time performance function and construct an error transformation mechanism to convert the original tracking error constrained by the predetermined time performance into an unconstrained transformation error. Step S3: Construct the enhanced variable frequency Nussbaum function, establish the Nussbaum function state update mechanism and the corresponding control quantity generation mechanism; Step S4: Construct a recursive control structure based on the dynamic surface inversion idea of ​​command filter, and use a switching neural network to approximate the unknown nonlinear time-varying dynamics online; Step S5: Combine the transformation error, Nussbaum compensation term, filter compensation term, and switch the neural network output to construct the actual control surface input and apply it to the deformable aircraft.

[0007] Preferably, in step S1, the dynamic model of the time-varying nonlinear system is: ; in, It is a time variable; The dimension of the state vector is a positive integer; Let be the system state vector. Indicates transpose; For control input; For system output; take respectively and For subsequent use of the index, where Used before expression State equations, Used to provide a unified description of the system state and variables of each channel; for any , Represents state components Regarding time The first derivative, The control coefficients are unknown time-varying coefficients with unknown signs, and there are unknown positive constants. and , so that for any All ; For unknown time-varying parameter vectors; It is an unknown smooth nonlinear function; For unknown but bounded external disturbances; System tracking error for: ; in, Given a smooth, bounded reference instruction trajectory.

[0008] Preferably, for the longitudinal attitude control application scenario of deformable aircraft, the system state vector and control input are as follows: ; ; in, The flight path angle, For the angle of attack, for Pitch angular velocity, This refers to the rudder deflection angle.

[0009] Preferably, in step S2, a predetermined time performance function is constructed. for: ; in, The pre-set convergence time, These are the preset steady-state performance parameters.

[0010] Preferably, in step S2, the error transformation mechanism is implemented through the following transformation function: ; ; in, The unconstrained error variable after error transformation; To address system tracking errors The normalized mapping function, The design parameters are used; through the above error transformation, the original constrained system tracking error is transformed. Transform into unconstrained error variables ; Define the comprehensive error signals at each level as follows: ; in, For hierarchical indexes, For the system number Each state component and Representing level 1 and level 2 respectively Level integrated error signal, and This is the signal for filtering error compensation. For the first The output signal of the stage command filter.

[0011] Preferably, in step S3, the enhanced frequency converter Nussbaum function is: ; in, For the first Enhanced frequency conversion Nussbaum function for each channel. It is a natural constant. and For design parameters, It serves as the independent variable of the Nussbaum function and is updated online as a state variable; Establish the state update mechanism for the Nussbaum function and the corresponding first... The generation mechanism of level control variables is as follows: For the first Level channel, construct intermediate control quantity And obtain the first number through the Nussbaum function. Level virtual control quantity At the same time, update the state of the Nussbaum function. ,satisfy: ; ; ; ; in, For the first Level integrated error signal; For the first Level filtering error compensation signal; For the first Level basis function vector; Unknown upper bound constant The estimated value, For estimating parameters The time derivative; State of the Nussbaum function The time derivative; It is a non-negative function; It is a positive integrable time-varying function; All of these are design parameters.

[0012] Preferably, the dynamic equation of the command filter is: ; in, For command filter input, For the output of the command filter, express Regarding time The first derivative, The filtering time constant; To compensate for the error introduced by the command filter, a compensation signal is constructed. Its renewal law is: ; in, express Regarding time The first derivative.

[0013] Preferably, in step S4, a switching neural network is used to approximate the unknown nonlinear time-varying dynamics online, and the approximation model is: ; in, For the first An unknown nonlinear time-varying dynamic term The online estimate, For the first Level network input vector, Given a basis function vector, For the weight estimation vector, This is the switching function; The adaptive update law for switching neural network weights is: ; in, Weight estimation vector Regarding time The first derivative, The learning rate matrix is ​​positive definite. For the first Level integrated error signal.

[0014] Preferably, in step S5, the actual control input for the control surfaces is constructed. , of which The virtual control quantity and the actual control input satisfy the following: ; ; ; ; ; in, For the first Intermediate control quantity at level The actual control input applied to the control surfaces of the deformable aircraft; For the first Level integrated error signal; This is the signal for filtering error compensation. Indicates its relation to time The first derivative; For the first Level basis function vector; Unknown upper bound constant The estimated value, for ） Regarding time The first derivative; State of the Nussbaum function Regarding time The derivative; It is a positive integrable time-varying function; A nonnegative function that satisfies the integrability condition; These are design parameters.

[0015] The present invention also provides a predetermined time tracking control system for a deformable aircraft with unknown control coefficients, for implementing the predetermined time tracking control method for a deformable aircraft with unknown control coefficients as described above, the system comprising: The status and command acquisition module is used to acquire measurable state quantities and reference command trajectories in the longitudinal attitude channel of the deformable aircraft. The predetermined time error transformation module, connected to the status and instruction acquisition module, is used to introduce a predetermined time performance function and construct an error transformation mechanism to convert the original tracking error constrained by the predetermined time performance into an unconstrained transformation error. An unknown direction adaptive compensation module is connected to a predetermined time error transformation module. It is used to construct an enhanced frequency conversion Nussbaum function, establish a Nussbaum function state update mechanism, and generate corresponding virtual control quantities to adaptively compensate for unknown control directions. The recursive control and uncertainty approximation module is connected to the unknown direction adaptive compensation module. The recursive control structure is constructed based on the dynamic surface inversion idea of ​​the command filter, and the switching neural network is used to approximate the unknown nonlinear time-varying dynamics in the system online. The control synthesis and output module is connected to the predetermined time error transformation module, the unknown direction adaptive compensation module, and the recursive control and uncertainty approximation module, respectively. It is used to synthesize the transformation error, Nussbaum compensation term, filter compensation term, and switch the neural network output to construct and output the actual control input applied to the control surface of the deformable aircraft.

[0016] Therefore, the present invention employs the aforementioned method and system for tracking and controlling deformable aircraft with unknown control coefficients at a predetermined time, and the beneficial technical effects are as follows: (1) Compared with traditional tracking control methods that asymptotically converge or have difficulty explicitly limiting the transition process, this invention introduces a predetermined time performance function and an error transformation mechanism to convert the constrained tracking error into an unconstrained transformation error, so that the tracking error converges within a predetermined time. It can enter and remain within a predefined performance region, thereby enabling preset and constrained tracking accuracy and convergence time; (2) Compared with existing control methods based on known control directions or requiring consistent control directions for each channel, this invention constructs an enhanced frequency conversion Nussbaum function and establishes a corresponding state update mechanism to achieve adaptive compensation for unknown control directions, avoids dependence on prior information of control directions, and thus improves the applicability of the method. (3) Compared with the conventional backstepping control method, which requires repeated differentiation of the virtual control quantity and is prone to complexity explosion, the present invention combines the dynamic surface inversion idea of ​​command filtering to construct a recursive control structure and introduces a filter error compensation term, which reduces the implementation complexity and improves the engineering feasibility while ensuring control performance. (4) The present invention uses a switching neural network to approximate and adaptively update the uncertainty of the nonlinear time-varying model generated during the deformation process, which is beneficial to maintain the stability and tracking performance of the control system under the conditions of drastic changes in aerodynamic parameters and the presence of external disturbances. Attached Figure Description

[0017] Figure 1 This is a flowchart of the predetermined time tracking control method for a deformable aircraft with unknown control coefficients according to the present invention. Figure 2 This is a schematic diagram of the longitudinal attitude dynamics structure of a deformable aircraft. Figure 3 This is a schematic diagram of the sweep angle command of a deformable aircraft changing over time. Figure 4 This is a schematic diagram of the closed-loop response curve in an embodiment of the present invention, wherein, Figure 4 In the diagram, (a) represents the closed-loop state and the reference trajectory. Figure 4 (b) in the figure represents the tracking error and the predetermined performance boundary (including a magnified view). Figure 4 (c) in the figure represents the actual control input curve; Figure 5 This is a schematic diagram of the internal adaptive variable change curve in an embodiment of the present invention, wherein, Figure 5 (a) in the diagram represents the output of the enhanced frequency conversion Nussbaum function. Figure 5 (b) in the figure represents the estimated value of the adaptive parameter. Figure 5 (c) in the text represents the Nussbaum state variable. Figure 5 In the figure, (d) is the neural network weight estimation curve. Detailed Implementation

[0018] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0019] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0020] Example 1 like Figure 1 As shown, the present invention provides a predetermined time tracking control method for a deformable aircraft with unknown control coefficients, comprising the following steps: In this invention, the deformable aircraft considered is a type of variable-sweep wing aircraft, whose longitudinal attitude dynamics structure is as follows: Figure 2 As shown. Figure 2 The process of sweep angle shape change of the deformable aircraft is presented. As the sweep angle command changes with the mission phase, aerodynamic parameters such as wing sweep angle, lift coefficient, and moment coefficient change with time, resulting in strong nonlinearity and significant time-varying parameter characteristics in the longitudinal attitude dynamics.

[0021] like Figure 3 As shown, the sweep angle command is a piecewise variation curve: it maintains a certain initial sweep angle in the initial stage, smoothly changes to the target sweep angle in the middle stage according to a predetermined law, and maintains a new constant value in the final stage; the corresponding deformation rate is a bounded function that varies with time, and its definition is: ; in, This is the current sweep angle command. and These represent the minimum and maximum sweep angles allowed during the mission, respectively. Therefore, when the sweep angle is within the interval... Deformation rate during internal changes Normalized to The sweep angle is used to characterize the relative progress of wing deformation. Changes in the sweep angle directly affect the position of the aerodynamic center and the distribution of lift and pitch moments, thereby altering the equivalent dynamics of the longitudinal attitude channel.

[0022] Based on the above physical background, this invention proposes a predetermined time tracking control method for deformable aircraft with unknown control coefficients (e.g., Figure 1 The specific steps are as follows: Step S1: Obtain the measurable state variables and reference command trajectory of the longitudinal attitude channel of the deformable aircraft. Based on the characteristics of the aerodynamic parameters changing with time during the deformation process of the deformable aircraft, the longitudinal dynamics are abstracted into a time-varying nonlinear system containing model uncertainty, external disturbances and multiple unknown time-varying control coefficients.

[0023] Based on the characteristics of aerodynamic parameters changing with time during the configuration deformation of deformable aircraft, the longitudinal attitude dynamics of deformable aircraft are abstracted into a time-varying nonlinear system containing model uncertainties, external disturbances, and multiple unknown time-varying control coefficients. Its dynamic model is described as follows: ; in, It is a time variable; The dimension of the state vector is a positive integer; Let be the system state vector. Indicates transpose; For control input; For system output; take respectively and For subsequent use of the index, where Used before expression State equations, Used to provide a unified description of the system state and variables of each channel; for any , Represents state components Regarding time The first derivative, The control coefficients are unknown time-varying coefficients with unknown signs, and there are unknown positive constants. and , so that for any All ; For unknown time-varying parameter vectors; It is an unknown smooth nonlinear function; For unknown but bounded external disturbances.

[0024] To achieve the reference command trajectory Tracking control, defining the system tracking error for: ; in, Given a smooth, bounded reference instruction trajectory.

[0025] For the longitudinal attitude control application scenario of deformable aircraft, the system state vector and control input are selected as follows: ; ; in, The flight path angle, For the angle of attack, The pitch angular velocity, For rudder deflection; The corresponding longitudinal dynamic equation is written as: ; in, , , They represent , , The first derivative with respect to time Let be the deformation rate defined by the change in the sweep angle. The nonlinear function can be specifically chosen as... , , , , To balance flight speed, It is the acceleration due to gravity. , , , , For deformation rate The varying aerodynamic derivative function is fitted using a polynomial form; It represents the unknown bounded disturbance term caused by multi-source uncertainties and external disturbances.

[0026] Step S2: Introduce a predetermined time performance function and construct an error transformation mechanism to equivalently transform the original tracking error constrained by the predetermined time performance into an unconstrained transformed error, so that the tracking error converges within a predetermined time. It enters a predefined performance region and remains within that region thereafter.

[0027] Construct a predetermined time performance function Its definition is: ; in, The pre-set convergence time, These are preset steady-state performance parameters used to limit the steady-state accuracy of the system tracking error after a predetermined time.

[0028] Define the error transformation function as follows: ; ; in, The unconstrained error variable after error transformation; To address system tracking errors The normalized mapping function, The design parameters are used; through the above error transformation, the original constrained system tracking error is transformed. Transform into unconstrained error variables .

[0029] Further define the comprehensive error signals at each level as follows: ; in, For hierarchical indexes, For the system number Each state component and Representing level 1 and level 2 respectively Level integrated error signal, and This is the signal for filtering error compensation. For the first The output signal of the stage command filter. By stabilizing and controlling the comprehensive error signals at each stage, the transformation error is ensured. Convergence, thus reducing the original tracking error At the predetermined convergence time It enters and remains within the predefined performance area.

[0030] Step S3: To address the issue of multiple unknown control coefficients in time-varying nonlinear systems and the problem that these control coefficients may be unknown and inconsistent, an enhanced variable frequency Nussbaum function with different frequency parameters is constructed for different control channels. A corresponding Nussbaum state update mechanism is established, and the Nussbaum function is introduced into the control law to achieve adaptive compensation for unknown control directions, thereby avoiding dependence on prior information about the control direction.

[0031] An enhanced frequency converter Nussbaum function with different frequency parameters is constructed for different control channels, and its definition is: ; in, For the first Enhanced frequency conversion Nussbaum function for each channel. It is a natural constant. and For design parameters, As the independent variable of the Nussbaum function and used as a state variable for online updates; by setting different values ​​for different control channels This enables adaptive compensation for multiple unknown control directions and their potential inconsistencies.

[0032] Establish the state update mechanism for the Nussbaum function and the corresponding first... The level control quantity generation mechanism is as follows: For the first Level channel, construct intermediate control quantity And obtain the first number through the Nussbaum function. Level virtual control quantity At the same time, update the state of the Nussbaum function. ,satisfy: ; ; ; ; in, For the first Level integrated error signal; For the first Level filtering error compensation signal; For the first Level basis function vector; Unknown upper bound constant The estimated value, For estimating parameters The time derivative; State of the Nussbaum function The time derivative; It is a non-negative function; It is a positive integrable time-varying function; All of these are design parameters.

[0033] Step S4: Construct a recursive control structure based on the dynamic surface inversion idea of ​​command filtering, set command filters for the step-by-step virtual control law, and use the filter output to replace the derivative of the virtual control signal to avoid repeated differentiation in the backstep design and improve the feasibility of the control method; at the same time, in view of the aerodynamic parameter changes caused by configuration deformation and the uncertainty of nonlinear time-varying model, construct a switching neural network to approximate the unknown nonlinear terms online, and design an adaptive update law for the neural network weights.

[0034] The command filter is set for the step-by-step virtual control law, and its filtering dynamic equation is: ; in, For command filter input, For the output of the command filter, express Regarding time The first derivative, The filtering time constant is used to replace the virtual control derivative and avoid repeated differentiation in backstep design.

[0035] To compensate for the error introduced by the command filter, a compensation signal is constructed. Its renewal law is: ; in, express Regarding time The first derivative.

[0036] A switching neural network is constructed to approximate the unknown nonlinear term online. The approximation model is as follows: ; in, For the first An unknown nonlinear time-varying dynamic term The online estimate, For the first Level network input vector, Given a basis function vector, For the weight estimation vector, This is the switching function; The adaptive update law for switching neural network weights is: ; in, Weight estimation vector Regarding time The first derivative, The learning rate matrix is ​​positive definite. For the first Level integrated error signal.

[0037] Step S5: Combine the transformation error, Nussbaum compensation term, filter compensation term and neural network output obtained from steps S2 to S4 to construct the actual control surface input and apply it to the deformable aircraft. This enables the closed-loop system to meet the predetermined time performance constraints under external disturbances and configuration deformation conditions, achieve predetermined time tracking of the reference command trajectory, and ensure that all signals of the closed-loop system are bounded.

[0038] Construct the actual control input for the control surface, where the first... The virtual control quantity and the actual control input satisfy the following: ; ; ; ; ; in, For the first Intermediate control quantity at level The actual control input applied to the control surfaces of the deformable aircraft; For the first Level integrated error signal; This is the signal for filtering error compensation. Indicates its relation to time The first derivative; For the first Level basis function vector; Unknown upper bound constant The estimated value, for ） Regarding time The first derivative; State of the Nussbaum function Regarding time The derivative; It is a positive integrable time-varying function; A nonnegative function that satisfies the integrability condition; These are design parameters.

[0039] The invention will be further illustrated below with specific examples.

[0040] Based on the predetermined time tracking control method for deformable aircraft with unknown control coefficients of the present invention, numerical simulation of the longitudinal attitude control of the deformable aircraft is performed using the MATLAB simulation platform, as follows: A longitudinal attitude dynamics model is established using a certain type of variable-sweep wing deformable aircraft as the object. Based on the characteristics of aerodynamic parameters changing with time during deformation, it is abstracted into a time-varying nonlinear system containing model uncertainties, external disturbances, and multiple unknown time-varying control coefficients. The control directions of each control channel are unknown and may be inconsistent. Simulation parameters are taken as follows: total mass of the aircraft. Balanced flight speed Gravitational acceleration External disturbances are set to , , .

[0041] To characterize the impact of configuration deformation on attitude dynamics, the sweep angle deformation process and its time-varying effect on aerodynamic parameters are considered in the simulation. The deformation rate is assumed to be... The value smoothly changes from 0 to 0.3 within the interval, and then remains unchanged.

[0042] In this embodiment, the reference trajectory is set to... The initial conditions are set as follows: system state. Adaptive parameter estimation initial value Nussbaum initial state Initial values ​​of neural network weights Regarding controller parameter settings: Enhanced frequency converter Nussbaum function parameter settings... , , Frequency parameters are taken , , This is used for adaptive compensation of unknown control coefficients and unknown control directions across multiple channels. The predetermined time performance index is taken as the predetermined convergence time. Steady-state performance parameters Error normalization parameter , and select , , , , , , , , , , ,as well as .

[0043] Figure 4 This is a schematic diagram of the closed-loop response curve, where: Figure 4 (a) in the figure gives the response relationship between the closed-loop state and the reference trajectory; Figure 4 Figure (b) shows the tracking error and the predetermined performance boundary (including a magnified view). It can be seen that the tracking error is within the predetermined convergence time. It enters the predetermined performance region and remains within the predetermined performance boundary in subsequent periods (including the period affected by time-varying disturbances caused by configuration deformation); Figure 4 (c) in the figure gives the actual control input curve. The control input shows transient adjustment in the initial stage, then quickly stabilizes and remains bounded. It only produces small adjustments near the deformation stage and does not cause the error to exceed the limit.

[0044] Figure 5 This is a schematic diagram of the internal adaptive variable variation curve, where: Figure 5 (a) Output of the enhanced frequency conversion Nussbaum function; Figure 5 (b) in the figure represents the estimated value of the adaptive parameter; Figure 5 (c) represents the Nussbaum state variables; Figure 5 In the diagram, (d) represents the neural network weight estimation curve. (From...) Figure 5 It can be seen that each internal adaptive variable remains bounded as it evolves over time, and the neural network weight estimation shows a convergent trend, thus verifying the effectiveness of the method of the present invention in achieving predetermined time tracking under conditions of unknown control direction, multi-source uncertainty, and configuration deformation.

[0045] It is worth noting that all contents not described in detail in this invention are existing technologies and are well known to those skilled in the art.

[0046] Therefore, the present invention employs the aforementioned predetermined time tracking control method and system for deformable aircraft with unknown control coefficients. It can adaptively compensate for unknown control directions without prior information on the control direction. At the same time, it can ensure that all signals of the closed-loop system are bounded and meet the predetermined time performance requirements under the conditions of configuration deformation and external disturbances. The present invention provides technical support for high-performance predetermined time tracking control of complex nonlinear time-varying systems in the future, and is especially suitable for high-precision and fast attitude tracking control tasks of deformable aircraft and other high-tech equipment.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for time-tracking control of a deformable aircraft with unknown control coefficients, characterized in that, Includes the following steps: Step S1: Obtain the measurable state variables and reference command trajectory of the longitudinal attitude channel of the deformable aircraft. Based on the time-varying characteristics of aerodynamic parameters, the longitudinal dynamics are abstracted into a time-varying nonlinear system with unknown time-varying control coefficients, and the system tracking error is defined. Step S2: Introduce a predetermined time performance function and construct an error transformation mechanism to convert the original tracking error constrained by the predetermined time performance into an unconstrained transformation error. Step S3: Construct the enhanced variable frequency Nussbaum function, establish the Nussbaum function state update mechanism and the corresponding control quantity generation mechanism; Step S4: Construct a recursive control structure based on the dynamic surface inversion idea of ​​command filter, and use a switching neural network to approximate the unknown nonlinear time-varying dynamics online; Step S5: Combine the transformation error, Nussbaum compensation term, filter compensation term, and switch the neural network output to construct the actual control surface input and apply it to the deformable aircraft.

2. The predetermined time tracking control method for a deformable aircraft with unknown control coefficients according to claim 1, characterized in that, In step S1, the dynamic model of the time-varying nonlinear system is: ； in, It is a time variable; The dimension of the state vector is a positive integer; Let be the system state vector. Indicates transpose; For control input; For system output; take respectively and For subsequent use of the index, where Used before expression State equations, Used to provide a unified description of the system state and variables of each channel; for any , Represents state components Regarding time The first derivative, The control coefficients are unknown time-varying coefficients with unknown signs, and there are unknown positive constants. and , so that for any All ; For unknown time-varying parameter vectors; It is an unknown smooth nonlinear function; For unknown but bounded external disturbances; System tracking error for: ； in, Given a smooth, bounded reference instruction trajectory.

3. The predetermined time tracking control method for a deformable aircraft with unknown control coefficients according to claim 2, characterized in that, For the longitudinal attitude control application scenario of deformable aircraft, the system state vector and control input are as follows: ； ； in, The flight path angle, For the angle of attack, for Pitch angular velocity, This refers to the rudder deflection angle.

4. The predetermined time tracking control method for a deformable aircraft with unknown control coefficients according to claim 3, characterized in that, In step S2, a predetermined time performance function is constructed. for: ； in, The pre-set convergence time, These are the preset steady-state performance parameters.

5. The predetermined time tracking control method for a deformable aircraft with unknown control coefficients according to claim 4, characterized in that, In step S2, the error transformation mechanism is implemented through the following transformation function: ； ； in, The unconstrained error variable after error transformation; To address system tracking errors The normalized mapping function, The design parameters are used; through the above error transformation, the original constrained system tracking error is transformed. Transform into unconstrained error variables ; Define the comprehensive error signals at each level as follows: ； in, For hierarchical indexes, For the system number Each state component and Representing level 1 and level 2 respectively Level integrated error signal, and This is the signal for filtering error compensation. For the first The output signal of the stage command filter.

6. The predetermined time tracking control method for a deformable aircraft with unknown control coefficients according to claim 5, characterized in that, In step S3, the enhanced inverter Nussbaum function is: ； in, For the first Enhanced frequency conversion Nussbaum function for each channel. It is a natural constant. and For design parameters, It serves as the independent variable of the Nussbaum function and is updated online as a state variable; Establish the state update mechanism for the Nussbaum function and the corresponding first... The generation mechanism of level control variables is as follows: For the first Level channel, construct intermediate control quantity And obtain the first number through the Nussbaum function. Level virtual control quantity At the same time, update the state of the Nussbaum function. ,satisfy: ； ； ； ； in, For the first Level integrated error signal; For the first Level filtering error compensation signal; For the first Level basis function vector; Unknown upper bound constant The estimated value, For estimating parameters The time derivative; State of the Nussbaum function The time derivative; It is a non-negative function; It is a positive integrable time-varying function; All of these are design parameters.

7. The predetermined time tracking control method for a deformable aircraft with unknown control coefficients according to claim 6, characterized in that, The dynamic equation of the command filter is: ； in, For command filter input, For the output of the command filter, express Regarding time The first derivative, The filtering time constant; To compensate for the error introduced by the command filter, a compensation signal is constructed. Its renewal law is: ； in, express Regarding time The first derivative.

8. The predetermined time tracking control method for a deformable aircraft with unknown control coefficients according to claim 7, characterized in that, In step S4, a switching neural network is used to approximate the unknown nonlinear time-varying dynamics online. The approximation model is as follows: ； in, For the first An unknown nonlinear time-varying dynamic term The online estimate, For the first Level network input vector, Given a basis function vector, For the weight estimation vector, This is the switching function; The adaptive update law for switching neural network weights is: ； in, Weight estimation vector Regarding time The first derivative, The learning rate matrix is ​​positive definite. For the first Level integrated error signal.

9. The predetermined time tracking control method for a deformable aircraft with unknown control coefficients according to claim 8, characterized in that, In step S5, the actual control input for the control surfaces is constructed. , of which The virtual control quantity and the actual control input satisfy the following: ； ； ； ； ； in, For the first Intermediate control quantity at level The actual control input applied to the control surfaces of the deformable aircraft; For the first Level integrated error signal; This is the signal for filtering error compensation. Indicates its relation to time The first derivative; For the first Level basis function vector; Unknown upper bound constant The estimated value, for ） Regarding time The first derivative; State of the Nussbaum function Regarding time The derivative; It is a positive integrable time-varying function; A nonnegative function that satisfies the integrability condition; These are design parameters.

10. A time-tracking control system for a deformable aircraft with unknown control coefficients, characterized in that, For implementing a predetermined time tracking control method for a deformable aircraft with unknown control coefficients as described in any one of claims 1-9, the system comprises: The status and command acquisition module is used to acquire measurable state quantities and reference command trajectories in the longitudinal attitude channel of the deformable aircraft. The predetermined time error transformation module, connected to the status and instruction acquisition module, is used to introduce a predetermined time performance function and construct an error transformation mechanism to convert the original tracking error constrained by the predetermined time performance into an unconstrained transformation error. An unknown direction adaptive compensation module is connected to a predetermined time error transformation module. It is used to construct an enhanced frequency conversion Nussbaum function, establish a Nussbaum function state update mechanism, and generate corresponding virtual control quantities to adaptively compensate for unknown control directions. The recursive control and uncertainty approximation module is connected to the unknown direction adaptive compensation module. The recursive control structure is constructed based on the dynamic surface inversion idea of ​​the command filter, and the switching neural network is used to approximate the unknown nonlinear time-varying dynamics in the system online. The control synthesis and output module is connected to the predetermined time error transformation module, the unknown direction adaptive compensation module, and the recursive control and uncertainty approximation module, respectively. It is used to synthesize the transformation error, Nussbaum compensation term, filter compensation term, and switch the neural network output to construct and output the actual control input applied to the control surface of the deformable aircraft.

Images