A flexible preset performance control method for waverider aircraft

Abstract

A flexible preset performance control method for a waverider aircraft comprises the following steps: constructing a motion model of the waverider aircraft; using a neural network to approximate a system function of a subsystem; constructing a first flexible performance function, and constraining a velocity tracking error based on the first flexible performance function; correcting a constraint transformation based on a velocity compensation system; constructing an expected value of a control input of a velocity subsystem motion model based on the corrected constraint transformation; determining the control input of the velocity subsystem motion model; constructing a second flexible performance function, and constraining an altitude tracking error based on the second flexible performance function; defining a conversion error of the altitude tracking error and a reference instruction of a track angle; constructing an expected value of a control input of a altitude subsystem motion model; determining the control input of the altitude subsystem motion model; and tracking and controlling the waverider aircraft according to the velocity subsystem motion model and the altitude subsystem motion model.

Description

Technical Field

[0001] The present invention belongs to the technical field of aircraft control, and in particular relates to a flexible preset performance control method for an actuator saturation constrained waverider aircraft. Background Art

[0002] Waverider Vehicle (WV) performs large maneuvering flights in near space, which places extremely high demands on the dynamic performance and steady-state accuracy of its control system. The Prescribed Performance Control (PPC) method can ensure that the WV control system has the desired dynamic performance and steady-state accuracy. However, the existing Prescribed Performance Control method, such as the Prescribed Performance Control method disclosed in the Chinese Invention Patent Application No. 2021110210608, has a rigid constraint envelope and no flexible adjustment capability. When there is an actuator saturation constraint, the tracking error will increase significantly. When the tracking error increases to approach or even exceed the constraint envelope, it will lead to control singularity and control system failure. Summary of the invention

[0003] The purpose of the present invention is to provide a flexible preset performance control method for a waverider aircraft, which can solve the technical defect that the existing PPC method cannot handle the actuator saturation constraint problem.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] A method for controlling the flexible preset performance of a waverider aircraft comprises the following steps:

[0006] S1. Constructing a motion model of the waverider aircraft, wherein the motion model of the waverider aircraft includes a velocity subsystem motion model and an altitude subsystem motion model;

[0007] S2, using a neural network to approximate the system function of the speed subsystem and the system function of the height subsystem;

[0008] S3, constructing a first flexible performance function for the speed tracking error, and performing a flexible envelope constraint on the speed tracking error based on the first flexible performance function;

[0009] S4, correcting the constraint transformation based on the speed compensation system to obtain a corrected constraint transformation;

[0010] S5. constructing the expected value of the control input of the velocity subsystem motion model based on the modified constraint transformation;

[0011] S6, determining the control input of the speed subsystem motion model according to the expected value of the control input of the speed subsystem motion model;

[0012] S7, constructing a second flexible performance function for the height tracking error, and performing a flexible envelope constraint on the height tracking error based on the second flexible performance function;

[0013] S8, defining the conversion error of the altitude tracking error and the reference instruction of the track angle;

[0014] S9, constructing the expected value of the control input of the height subsystem motion model;

[0015] S10, determining the control input of the altitude subsystem motion model according to the expected value of the control input of the altitude subsystem motion model;

[0016] S11. Tracking and controlling the waverider aircraft according to the velocity subsystem motion model and the altitude subsystem motion model.

[0017] In the above method, optionally, in step S1, the velocity subsystem motion model is:

[0018] In the formula is the first-order derivative of V with respect to time t, V is the flight speed of the waverider aircraft, u V is the control input of the velocity subsystem motion model, ξ V is the system function of the speed subsystem;

[0019] The altitude subsystem motion model is:

[0020]

[0021] In the formula is the state of the height subsystem, is the system function of the height subsystem, u h It is the control input of the altitude subsystem.

[0022] In the above method, optionally, in step S2,

[0023] Approximate the system function of the velocity subsystem: In the formula, ω ξ,V is the first weight vector of the neural network, β ξ,V (V) is the first basis function vector of the neural network, ι ξ,V is the first approximation error of the neural network;

[0024] Approximate the system function of the height subsystem:

[0025] In the formula, ω ξ,his the second weight vector of the neural network, is the second basis function vector of the neural network, ι ξ,h is the second approximation error of the neural network.

[0026] According to the method described above, optionally, in step S3, the speed tracking error s V (t) is: s V (t) = VV s , V s It is the reference instruction of the speed subsystem;

[0027] The first performance function is: In the formula sign(·) is the sign function, s V (0) is s V (t) is the value at t = 0,

[0028] d V,L d V,R Satisfying 0<d V,L <1,0<d V,R <1, p V,0 is the initial value of the constraint envelope of the velocity tracking error, p V,T is the final value of the constraint envelope of the velocity tracking error, T V is the velocity tracking error convergence time, σ V is the convergence smoothing coefficient, is the first flexible adjustment item;

[0029] Will As s V The lower envelope of (t) will be As s V The upper envelope of (t) is:

[0030] According to the method described above, optionally, in step S4, the constraint transformation e after correction is performed s,V =ε s,V -x V , x V is the speed compensation system, ε s,V is the constraint transformation,

[0031] According to the method described above, optionally, in step S5, the expected value of the control input of the speed subsystem motion model is Among them, k e,V,1 is the response speed coefficient of constraint transformation, k e,V,2 is the response integral coefficient of the constraint transformation, V sThe first derivative with respect to time t is is the estimate of the norm of the first weight vector of the neural network, Δ x,V and a x,V,1 is the design parameter of the first flexibility adjustment item;

[0032] In step S6, the control input of the speed subsystem motion model In the formula and u V The allowed lower and upper bounds.

[0033] According to the method described above, optionally, in step S7, the height tracking error s h (t) is: s h (t) = hh s ,h s It is the reference command of the altitude subsystem;

[0034] The second performance function is: In the formula s h (0) is s h (t) is the value at t = 0,

[0035] d h,L d h,R Satisfying 0<d h,L <1,0<d h,R <1, p h,0 is the initial value of the height tracking error constraint envelope, p h,T is the final value of the height tracking error constraint envelope, T h is the height tracking error convergence time, σ h is the height tracking error convergence smoothing coefficient, is the second flexible adjustment item;

[0036] Will As s h The lower envelope of (t) will be As s h The upper envelope of (t) is:

[0037] In the above method, optionally, in step S8,

[0038] The conversion error of height tracking error is:

[0039] The reference instructions for the track angle are: k in the formula h,γ is the design parameter of the reference instruction of the track angle, is the first-order derivative of the reference command of the altitude subsystem with respect to time t,

[0040] According to the method described above, optionally, in step S9, the expected value of the control input of the height subsystem motion model is

[0041]

[0042] Among them, S γ is the error function, E γ To correct the tracking error, σ γ To correct the tracking error convergence coefficient, k s,h is the error function convergence coefficient, For γ s The third-order derivative with respect to time t, is the second weight vector ω of the neural network ξ,h The norm θ ξ,h The estimated height compensation system is:

[0043]

[0044] In step S10, the control input of the height subsystem motion model In the formula and u h The allowed lower and upper bounds.

[0045] From the above technical solutions, it can be seen that the present invention designs a new constraint envelope with flexible adjustment capability, and a compensation system to compensate for the actuator saturation constraint, and uses the corrected conversion error for the feedback control law. and The design ensures that the tracking error is always within the constraint envelope when there is an actuator saturation constraint, thereby making up for the technical defect that the existing PPC method cannot handle the actuator saturation constraint problem. The designed constraint envelope can flexibly adjust its shape according to the actuator saturation constraint, so that the tracking error is always within the constraint envelope, avoiding the control singularity risk of traditional methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a flow chart of the method of the present invention;

[0047] Figure 2 It is a simulation diagram of the speed tracking effect of the method of the present invention;

[0048] Figure 3 It is a simulation diagram of the speed tracking error of the method of the present invention;

[0049] Figure 4 It is a simulation diagram of the height tracking effect of the method of the present invention;

[0050] Figure 5 It is a simulation diagram of the height tracking error of the method of the present invention;

[0051] Figure 6 A control input simulation diagram of the speed subsystem of the method of the present invention;

[0052] Figure 7 A control input simulation diagram of the altitude subsystem of the method of the present invention;

[0053] Figure 8 It is a simulation diagram of speed tracking error using the existing method;

[0054] Fig. 9 This is a simulation diagram of the height tracking error using the existing method.

[0055] The specific implementation modes of the present invention are further described in detail below with reference to the accompanying drawings. DETAILED DESCRIPTION

[0056] The present invention is described in detail below in conjunction with the accompanying drawings. When describing the embodiments of the present invention in detail, for the convenience of explanation, the drawings representing the device structure will not be partially enlarged according to the general proportion, and the schematic diagram is only an example, which should not limit the scope of protection of the present invention. It should be noted that the drawings are simplified in form and use non-precise proportions, which are only used to facilitate and clearly assist in explaining the purpose of the embodiments of the present invention. At the same time, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated; the terms "positive", "negative", "bottom", "upper", "lower", etc. indicate the orientation or position relationship based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation to the present invention.

[0057] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, it can also be the internal connection of two elements, it can be a wireless connection, or it can be a wired connection. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0058] Actuator saturation is a nonlinear phenomenon in control systems. Due to physical limitations, controller signals are limited to a specific range when transmitted to the actuator. When there is an actuator saturation constraint in the control system, the controller output signal is limited to the saturation range. For the preset performance control method of the rigid constraint envelope, the tracking error will increase significantly when there is an actuator saturation constraint. When the tracking error approaches or even exceeds the constraint envelope, it will cause control singularity and control system failure. To this end, the method of the present invention designs a new constraint envelope with flexible adjustment capability, and a compensation system to compensate for the actuator saturation constraint to solve the problem of control singularity and control system failure.

[0059] Figure 1 The flowchart of the method of the present invention is shown below. Figure 1 , the method of the present invention is further described, such as Figure 1 As shown, the method of the present invention comprises the following steps:

[0060] S1. Constructing a motion model of a waverider aircraft. The motion model of the waverider aircraft includes a velocity subsystem motion model and an altitude subsystem motion model.

[0061] The motion model of the speed subsystem is:

[0062] In the formula is the first-order derivative of V with respect to time t, V is the flight speed of the waverider aircraft, ξ V is the system function of the speed subsystem. In the subsequent steps, the present invention introduces a neural network to approximate the system function of the speed subsystem. Indicates partial derivative, u V It is the control input of the motion model of the speed subsystem;

[0063] The motion model of the altitude subsystem is:

[0064] In the formula is the state of the height subsystem, is the system function of the height subsystem. In the subsequent steps, the present invention introduces a neural network to approximate the system function of the height subsystem. u h is the control input of the altitude subsystem, γ is the track angle of the waverider vehicle, for The first derivative with respect to time t, and so on, for The first-order derivative with respect to time t is for The first derivative with respect to time t;

[0065] S2, using the same neural network to analyze the system function ξ of the speed subsystem V and the system function of the height subsystem Approximation is performed; the present invention does not limit the type of neural network used, and commonly known radial basis neural networks, inverse neural networks, etc. can be used;

[0066] Approximate the system function of the velocity subsystem: In the formula, ω ξ,V is the first weight vector of the neural network, β ξ,V (V) is the first basis function vector of the neural network, ι ξ,V is the first approximation error of the neural network, for ξ,V The upper bound of (·) T Represents the matrix transpose, where the first weight vector ω ξ,V In the subsequent steps, the update law is designed to update it. The control law and the update law do not require the first approximation error ι of the neural network. ξ,V and its upper bound The exact value of ι ξ,V and It is only used to describe the structure of the neural network approximation. ξ,V (V) is known, and its expression is determined according to the type of neural network selected;

[0067] For the system function of the height subsystem, In the formula, ω ξ,h is the second weight vector of the neural network, is the second basis function vector of the neural network, ι ξ,h is the second approximation error of the neural network, for ξ,h The upper bound of Among them, the second weight vector ω ξ,h In the subsequent steps, the update law is designed to update it. Neither the control law nor the update law requires the second approximation error ι of the neural network ξ,h and its upper bound The exact value of ι ξ,h and It is only used to describe the structure of the neural network approximation. is known, and its expression is determined according to the type of neural network selected; the present invention uses the same neural network to approximate the system function of the speed subsystem and the system function of the height subsystem, the parameters in the neural network are the same, and different subscripts are used for each parameter only to distinguish different system functions;

[0068] S3, speed tracking error s V (t) constructing a first flexibility performance function, and calculating the speed tracking error s based on the first flexibility performance function V (t) performing flexible constraint enveloping;

[0069] Speed ​​tracking error V (t) is: s V (t) = VV s , V s is the reference command of the speed subsystem, V s Given;

[0070] The first performance function is:

[0071] In the formula sign(·) is the sign function, s V (0) is s V (t) is the value at t = 0, d V,L d V,R Satisfying 0<d V,L <1,0<d V,R <1, p V,0 is the initial value of the velocity tracking error constraint envelope, p V,T is the final value of the velocity tracking error constraint envelope, when t = 0, p V (t) = p V,0 , t = T V Time V (t) = p V,T , p V,0 >p V,T >0, p V,0 With p V,T The value of is an empirical value, based on the speed tracking error s V The initial value of (t) and the expected size of the steady-state value are determined, that is, p V,0 To be better than s V The initial value of (t) is slightly larger, p V,T To be better than s V The final value of (t) is slightly larger, T V is the velocity tracking error convergence time, is the first flexible adjustment item; T V >0, which means p V (t) from pV,0 =p V (0) converges to p V,T =p V (T) The time required is selected according to actual needs, that is, if you want p V (t) converges faster, T V The value of is smaller. If you want p V (t) converges more slowly, T V The value of σ is larger, V is the velocity tracking error convergence smoothing coefficient, indicating p V (t) smoothness of convergence, σ V >0, its value is selected according to actual needs, that is, if p V (t) If the convergence is smoother, the value is larger, otherwise the value is smaller;

[0072] Will As s V The lower envelope of (t) will be As s V The upper envelope of (t) is:

[0073] First flexible adjustment b x,V,1 >0 is the first flexible adjustment item The gain coefficient is selected according to actual needs, that is, if If the amplitude is larger, the value is larger, otherwise the value is smaller. x,V,2 >0 is the first flexible adjustment item The response rate coefficient is selected according to actual needs, that is, if If the response is faster, the value is larger, otherwise the value is smaller; x V For the speed compensation system, For x V The first-order derivative with respect to time t is is the control input u of the velocity subsystem motion model V The expected value of x,V and a x,V,1 is the design parameter of the first flexibility adjustment term, Δ x,V and a x,V,1 Satisfy 0<Δ x,V <1, a x,V,1 >0, Δ x,V and a x,V,1 There is no specific range of values, depending on x V The specific convergence of VIf the convergence is faster, the value is larger, otherwise the value is smaller;

[0074] For the rigid constraint envelope in the existing preset performance control method, if the speed control actuator is in a saturated state, it is easy to cause control singularity problems; while the flexible constraint envelope of the present invention and Contains flexible adjustment items If the speed control actuator is in saturation, then At this time x V ≠0, and then the first flexible adjustment item To increase and reduce So that the constraint envelope and With flexible adjustment function, it ensures the speed tracking error s V (t) will not cross the constraint envelope This avoids the control singularity problem of the traditional preset performance control method; and the compensation system of the traditional PPC method is mostly constructed by unbounded functions, which easily causes the state of the compensation system to not converge. The speed compensation system of the present invention is constructed by bounded functions. It is constructed to ensure the stability of the speed compensation system and the state of the speed compensation system at all times. V Convergence of x V Convergence can be guaranteed It will not grow infinitely, thus avoiding control singularity problems;

[0075] S4, Constraint transformation ε based on speed compensation system s,V Make corrections and obtain the corrected constraint transformation e s,V ;

[0076] Corrected constraint transformation e s,V =ε s,V -x V , where ε s,V is the constraint transformation,

[0077] S5, based on the modified constraint transformation e s,V Construct the control input u of the velocity subsystem motion model V Expected value

[0078]

[0079] k e,V,1 is the response speed coefficient of the constraint transformation, affecting e s,V The response speed, k e,V,1 >0, its value should be reasonably selected according to actual needs, that is, if es,V If the response speed is faster, the value is larger, otherwise the value is smaller. e,V,2 is the integral coefficient of the constraint transformation, 1>k e,V,2 >0, V s The first derivative with respect to time t is is the norm of the first weight vector of the neural network θ ξ,V The estimate of θ ξ,V =||ω ξ,V || 2 , Δ ε,V =0.5, then,

[0080] Updates are made based on the following rules: k in the formula θ,V is the first gain coefficient, k θ,V >0, its value is based on The actual response situation can be reasonably selected. If the response speed is faster, then the value is larger, otherwise it is smaller; therefore, the neural network approximation in step S2 has only one adaptive parameter and the first weight vector ω in the existing method ξ,V Compared with the strategy of adaptively adjusting the elements, the amount of online learning can be reduced;

[0081] S6. Expected value of control input according to the speed subsystem motion model Determine the control input u of the velocity subsystem motion model V ;

[0082] In the formula and u V The allowed lower and upper bounds, u V The allowed lower and upper bounds are empirical values, as long as That's it;

[0083] S7, height tracking error s h (t) constructing a second flexibility performance function, and calculating the height tracking error s based on the second flexibility performance function h (t) Perform flexible constraint enveloping;

[0084] Height tracking error h (t) is: s h (t) = hh s ,h s is the given reference command of the altitude subsystem, h s Given;

[0085] The second performance function is:

[0086] In the formula s h (0) is s h (t) is the value at t = 0, d h,L d h,R Satisfying 0<d h,L <1,0<d h,R <1, is the second flexibility adjustment term, p h,0 is the initial value of the height tracking error constraint envelope, p h,T is the final value of the height tracking error constraint envelope, p at t = 0 h (t) = p h,0 , t = T h Time h (t) = p h,T , p h,0 >p h,T >0, T h is the height tracking error convergence time, p h,0 With p h,T The value of is determined according to the height tracking error s h The initial value of (t) and the expected size of the steady-state value are determined, that is, p h,0 To be better than s h The initial value of (t) is slightly larger, p h,T To be better than s h The final value of (t) is slightly larger, T h >0, which means p h (t) from p h,0 =p h (0) converges to p h,T =p h (T) The time required is selected according to actual needs, that is, if you want p h (t) converges faster, T h The value of is smaller. If you want p h (t) converges more slowly, T h The value of σ is larger, h is the height tracking error convergence smoothing coefficient, indicating p h (t) smoothness of convergence, σ h >0, its value is selected according to actual needs, that is, if p h (t) converges more smoothly and takes a larger value, otherwise it takes a smaller value;

[0087] Will As s h The lower envelope of (t) will be As s h The upper envelope of (t) is:

[0088] Second flexible adjustment item b x,h,1 >0 is the second flexible adjustment item The gain coefficient is selected according to actual needs, that is, if If the amplitude is larger, the value is larger, otherwise the value is smaller. x,h,21 , b x,h,22 , b x,h,23 The second flexible adjustment item The rate coefficient, b x,h,21 , b x,h,22 , b x,h,23 are all greater than zero, and the value is selected according to actual needs, that is, if necessary If the response is faster, the value is larger, otherwise the value is smaller; x h,1 、x h,2 、x h,3 Elements in the height compensation system;

[0089] The height compensation system is:

[0090]

[0091] In the formula, is the expected value of the control input of the height subsystem motion model, Δ x,h =0.5, then,

[0092] The compensation system in the existing traditional preset performance control method is mostly constructed by unbounded functions, which easily causes the state of the compensation system to not converge. The height compensation system of the present invention is constructed by bounded functions. Constructed to always ensure the stability of the height compensation system and x h,1 、x h,2 、x h,3 Convergence of , thus avoiding control singularity problems;

[0093] S8, define the conversion error ε of the height tracking error s,h and the reference command of the track angle γ s ;

[0094] Conversion error ε of height tracking error s,h for:

[0095]

[0096] Reference command of track angle γs :

[0097] k in the formula h,γ is the conversion error convergence coefficient, k h,γ >0, there is no specific range for its value, and it should be based on ε s,h The specific convergence situation of is the first-order derivative of the reference command of the altitude subsystem with respect to time t, for The first-order derivative with respect to time t is for The first derivative with respect to time t;

[0098] S9, construct the control input u of the height subsystem motion model h Expected value

[0099]

[0100] Among them, S γ is the error function, E γ To correct the tracking error, E γ =γ-γ s -x h,1 , σ γ To correct the tracking error convergence coefficient, σ γ >0, its value is selected according to actual needs, that is, if you want E γ If the convergence is faster, then the value is larger, otherwise the value is smaller. s,h is the error function convergence coefficient, k s,h >0, its value is selected according to actual needs, that is, if you want S γ If the convergence is faster, the value is larger, otherwise the value is smaller. For γ s The third-order derivative with respect to time t, is the second weight vector ω of the neural network ξ,h The norm θ ξ,h The estimate of θ ξ,h =||ω ξ,h || 2 ;

[0101] Updates are made based on the following rules: k in the formula θ,h is the second gain coefficient, k θ,h >0, its value is based on The actual response of the neural network is reasonably selected; therefore, the neural network approximation in step S2 has only one adaptive parameter Compared with the existing method for the second weight vector ω ξ,h Compared with the strategy of adaptively adjusting the elements, the amount of online learning can be reduced;

[0102] S10, expected value of control input according to the motion model of the altitude subsystem Determine the control input u of the altitude subsystem motion model h ;

[0103] In the formula and u h Allowed lower and upper bounds, and satisfy

[0104] S11. Tracking and controlling the waverider aircraft according to the velocity subsystem motion model and the altitude subsystem motion model.

[0105] The method of the present invention designs a flexible constraint envelope for the tracking error of the motion model of the velocity subsystem and the motion model of the height subsystem, breaking through the technical defect that the rigid constraint envelope of the traditional PPC method cannot autonomously adjust the constraint envelope boundary according to the error fluctuation situation, and designs a flexible adjustment item, which can use the state of the compensation system to perceive the error fluctuation situation, and autonomously adjust the upper and lower boundaries of the constraint envelope, avoiding the control singularity risk of the traditional PPC. At the same time, the present invention uses bounded functions to construct a compensation system to compensate for the saturation of the actuator, avoiding the problem of non-convergence of the compensation system state caused by the existing unbounded compensation, and uses the bounded state of the compensation system to correct the error, and adopts the correction error to design the expected control input, ensuring the stability of the closed-loop control system and the feasibility of the preset performance under the actuator saturation situation.

[0106] In order to verify the effect of the method of the present invention, the method of the present invention is used to track the speed and height reference input using Matab2021 software, and compared with the method disclosed in the Chinese invention patent application with application number 2021110210608 as the existing method. The simulation is solved using the fourth-order Runge-Kutta method, and the simulation step size is 0.01s. The simulation parameters are taken as: τ u,V =5, p V,0 =2.5, p V,T =0.3,σ V =4, b x,V,1 =2, b x,V,2 =3.5, a x,V,1 =1.5,Δ x,V =0.8,k e,V,1 =0.3,k e,V,2 =0.8,k θ,V=0.05,,p h,0 =0.7,p h,T =0.1,σ h =4,b x,h,1 =2,b x,h,21 =12,b x,h,22 =15,b x,h,23 =15,k h,γ =2,a x,h,1 =0.5,a x,h,2 =1,b x,h,3 =1,Δ x,h =0.8,σ γ =7,k s,h =80,k θ,h =0.5.

[0107] The simulation results are as follows Figures 2 to 7 As shown in the simulation results, the method of the present invention can ensure the stable tracking of the speed and height to the respective reference instructions, and the constraint envelope can autonomously adjust the constraint envelope shape according to the saturation of the control actuator, so that the speed tracking error and the height tracking error are always within their respective constraint envelopes, thereby meeting the expected preset performance.

[0108] Under the same circumstances, the existing method is used for comparative simulation. The simulation results are shown in Figure 8 and Fig. 9 As shown. Figure 8 and Fig. 9 It can be seen that, under the same conditions, since the constraint envelope of the existing method has no flexible adjustment term, when the tracking error gradually increases and approaches the boundary of the constraint envelope, it leads to control singularity; while under the same conditions, the method of the present invention can avoid the control singularity problem, thereby breaking through the technical defect that the existing method cannot effectively deal with the control actuator saturation problem, and ensuring the satisfactory preset performance of the tracking error under the actuator saturation condition.

[0109] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A flexible preset performance control method for a waverider aircraft, It is characterized in that The following steps are involved: S1. Constructing a motion model of the waverider aircraft, wherein the motion model of the waverider aircraft includes a velocity subsystem motion model and an altitude subsystem motion model; S2, using a neural network to approximate the system function of the speed subsystem and the system function of the height subsystem; S3, constructing a first flexible performance function for the speed tracking error, and performing a flexible constraint envelope on the speed tracking error based on the first flexible performance function; S4, correcting the constraint transformation based on the speed compensation system to obtain a corrected constraint transformation; S5. constructing the expected value of the control input of the velocity subsystem motion model based on the modified constraint transformation; S6, determining the control input of the speed subsystem motion model according to the expected value of the control input of the speed subsystem motion model; S7, constructing a second flexible performance function for the height tracking error, and performing a flexible constraint envelope on the height tracking error based on the second flexible performance function; S8, defining the conversion error of the altitude tracking error and the reference instruction of the track angle; S9, constructing the expected value of the control input of the height subsystem motion model; S10, determining the control input of the altitude subsystem motion model according to the expected value of the control input of the altitude subsystem motion model; S11. Tracking and controlling the waverider aircraft according to the velocity subsystem motion model and the altitude subsystem motion model.

2. The flexible preset performance control method of a waverider aircraft according to claim 1, Features: In step S1, the velocity subsystem motion model is: In the formula is the first-order derivative of V with respect to time t, V is the flight speed of the waverider aircraft, u V is the control input of the velocity subsystem motion model, ξ V is the system function of the speed subsystem; The altitude subsystem motion model is: In the formula is the state of the height subsystem, is the system function of the height subsystem, u h It is the control input of the altitude subsystem.

3. The flexible preset performance control method of a waverider aircraft according to claim 1, Features: In the step S2, Approximate the system function of the velocity subsystem: In the formula, ω ξ,V is the first weight vector of the neural network, β ξ,V (V) is the first basis function vector of the neural network, ι ξ,V is the first approximation error of the neural network; Approximate the system function of the height subsystem: In the formula, ω ξ,h is the second weight vector of the neural network, is the second basis function vector of the neural network, ι ξ,h is the second approximation error of the neural network.

4. The flexible preset performance control method of a waverider aircraft according to claim 1, Features: In step S3, the speed tracking error s V (t) is: s V (t) = VV s , V s It is the reference instruction of the speed subsystem; The first performance function is: In the formula sign(·) is the sign function, s V (0) is s V (t) is the value at t = 0, d V,L d V,R Satisfying 0<d V,L <1,0<d V,R <1, p V,0 is the initial value of the constraint envelope of the velocity tracking error, p V,T is the final value of the constraint envelope of the velocity tracking error, T V is the velocity tracking error convergence time, σ V is the convergence smoothing coefficient, is the first flexible adjustment item; Will As s V The lower envelope of (t) will be As s V The upper envelope of (t) is:

5. The flexible preset performance control method of a waverider aircraft according to claim 1, Features: In step S4, the modified constraint is transformed into: s,V =ε s,V -x V , x V is the speed compensation system, ε s,V is the constraint transformation, 6. The flexible preset performance control method of a waverider aircraft according to claim 1, Features: In step S5, the expected value of the control input of the speed subsystem motion model is Among them, k e,V,1 is the response speed coefficient of constraint transformation, k e,V,2 is the integral coefficient of the constraint transformation, V s The first derivative with respect to time t is is the estimate of the norm of the first weight vector of the neural network, Δ x,V and a x,V,1 is the design parameter of the first flexibility adjustment item; In step S6, the control input of the speed subsystem motion model In the formula and u V The allowed lower and upper bounds.

7. The flexible preset performance control method of a waverider aircraft according to claim 1, Features: In step S7, the height tracking error s h (t) is: s h (t) = hh s ,h s It is the reference command of the altitude subsystem; The second performance function is: In the formula s h (0) is s h (t) is the value at t = 0, d h,L d h,R Satisfying 0<d h,L <1,0<d h,R <1, p h,0 is the initial value of the height tracking error constraint envelope, p h,T is the final value of the height tracking error constraint envelope, T h is the height tracking error convergence time, σ h is the height tracking error convergence smoothing coefficient, is the second flexible adjustment item; Will As s h The lower envelope of (t) will be As s h The upper envelope of (t) is:

8. The flexible preset performance control method of a waverider aircraft according to claim 1, Features: In the step S8, Conversion error ε of height tracking error s,h for: The reference instructions for the track angle are: k in the formula h,γ is the design parameter of the reference instruction of the track angle, is the first-order derivative of the reference command of the altitude subsystem with respect to time t, 9. The quantitative preset performance control method of a waverider aircraft according to claim 1, Features: In step S9, the expected value of the control input of the height subsystem motion model Among them, S γ is the error function, E γ To correct the tracking error, σ γ To correct the tracking error convergence coefficient, k s,h is the error function convergence coefficient, For γ s The third-order derivative with respect to time t, is the second weight vector ω of the neural network ξ,h The estimate of the norm of the height compensation system is: In step S10, the control input of the height subsystem motion model In the formula and u h The allowed lower and upper bounds.

Images