A control method for morphing aircraft based on L1 adaptive dynamic inversion

Abstract

The present invention discloses a variant aircraft control method based on L1 adaptive dynamic inversion, which belongs to the field of variant aircraft flight control. The specific steps are as follows: establishing a six-degree-of-freedom nonlinear dynamic model of the variant aircraft; performing time-scale separation on the 12 state variables of the aircraft nonlinear dynamic model according to the singular perturbation theory; designing the angular velocity and attitude angle loop control laws based on the nonlinear dynamic inversion method according to the differential equations of the angular velocity and attitude angle loop of the variant aircraft; and designing the maneuver generator control law based on a method combining dynamic inversion and linear feedback with L1 adaptation. The present invention can improve the performance of the control system in the presence of uncertainty, improve the accuracy of tracking control, and improve the robustness of the system, so that the variant aircraft can still maintain stable flight under the conditions of input disturbance, aerodynamic parameter uncertainty, parameter changes, etc.

Description

Technical Field

[0001] The present invention belongs to the field of flight control of morphing aircraft, and in particular relates to a morphing aircraft control method based on L1 adaptive dynamic inversion. Background Art

[0002] A morphing aircraft is an aircraft that can change its appearance in response to changes in maneuverability, flight environment, or mission. In recent years, both military and civil aviation have placed demands on aircraft to perform a variety of missions under varying flight conditions, driven by new operational concepts and models. The application value and importance of morphing aircraft in both national defense, military, and civil aviation are self-evident, and have become a cutting-edge and hot topic of research in the international aerospace community.

[0003] Morphing aircraft differ from fixed-wing aircraft in that they incorporate a morphing mechanism. This leads to the following major technical challenges in flight control: 1) The distributed actuation characteristics of the morphing mechanism complicate modeling of the aircraft's dynamics; 2) Changes in the aircraft's structural shape lead to a series of issues, including changes in structural parameters and aerodynamic characteristics, which in turn place higher demands on flight control. Most studies on morphing aircraft flight control employ small-disturbance linearization of the nonlinear system, converting it approximately into a linear system, and then employing common linear system control methods to design control laws. However, this linearization process ignores higher-order nonlinear terms, significantly impacting control accuracy. Other studies simplify the six-degree-of-freedom dynamic model to three degrees of freedom, focusing solely on longitudinal control. Therefore, there is an urgent need to develop a nonlinear control scheme that can accurately track and adapt to changes in the aircraft's structural parameters and aerodynamic characteristics to achieve six-degree-of-freedom flight control for morphing aircraft. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In response to the shortcomings of the existing technology, the present invention provides a variant aircraft control method based on L1 adaptive dynamic inversion, which can achieve precise tracking and adapt to changes in aircraft structural parameters and aerodynamic characteristics, and meet the robustness requirements of variant aircraft flight control.

[0006] (2) Technical solution

[0007] In order to achieve the above technical objectives, the technical solution of the present invention includes the following steps:

[0008] A method for controlling a morphing aircraft based on L1 adaptive dynamic inversion comprises the following steps:

[0009] (1) Establish a nonlinear dynamic model of the morphing aircraft, which can characterize the dynamic characteristics of the aircraft during the morphing process;

[0010] (2) Based on the singular perturbation theory, the 12 state variables of the aircraft nonlinear dynamics model are time-scale separated;

[0011] (3) According to the differential equation of the angular velocity loop of the morphing aircraft, the angular velocity loop control law is designed based on the nonlinear dynamic inverse method;

[0012] (4) According to the differential equation of the attitude angle loop of the morphing aircraft, the attitude angle loop control law is designed based on the nonlinear dynamic inverse method;

[0013] (5) According to the differential equation of the maneuvering state quantity of the variant aircraft, the maneuvering generator control law is designed based on the linear feedback nonlinear dynamic inversion and L1 adaptive method.

[0014] Furthermore, in step (1), the nonlinear dynamic model of the variant aircraft includes:

[0015] A. Dynamic equations for the movement of the center of mass of an aircraft

[0016]

[0017]

[0018]

[0019] in, are the flight acceleration, angle of attack acceleration and sideslip acceleration of the aircraft respectively, V, α, β are the flight speed, angle of attack and sideslip angle of the aircraft respectively, m is the mass of the aircraft, p, q, r are the roll, pitch and yaw angular velocities of the aircraft respectively, F x ,F y ,F z are the components of the total external force acting on the aircraft in the body axis system, F ix ,F iy ,F iz are the components of the additional force generated by the aircraft due to the change in the center of mass position under the body axis system, and the calculation formula is:

[0020]

[0021] Among them, T is the engine thrust, D, Y, L are the drag, lift and side force of the aircraft respectively, g is the acceleration due to gravity, are the roll angle and pitch angle of the aircraft respectively, They are roll angular acceleration, pitch angular acceleration and yaw angular acceleration respectively; S x ,S y ,S z represents the static moment under the aircraft body axis, are the first-order derivative of the static moment with time, are the second derivatives of the static moment with time.

[0022] B. Dynamic equations for rotation around the center of mass

[0023] According to the theory of multi-rigid body dynamics, unlike conventional aircraft, a morphing aircraft will generate additional forces and torques due to deformation. The dynamic equation of the morphing aircraft rotating around its center of mass is deduced as follows:

[0024]

[0025] in, is the moment of inertia matrix of the aircraft, I x , I y , I z is the moment of inertia of the aircraft, I xy , I xz , I yz is the product of inertia. They are rolling moment, pitching moment and yaw moment respectively. They represent the components of flight acceleration on the body axis, and the expressions of b1, b2, and b3 are:

[0026]

[0027]

[0028]

[0029] Among them, u, v, w represent the components of the flight speed on the body axis, m i represents the weight of the i-th wing of the i-th aircraft, is the first derivative of the aircraft's moment of inertia with time, is the first derivative of the product of inertia with time, S ix ,S iy ,S iz Represent the static moment of the i-th wing of the aircraft, They represent the first-order derivative of the static moment of the i-th wing of the aircraft with time, They represent the second derivative of the static moment of the i-th wing of the aircraft with time.

[0030] C. Kinematic equations for the movement of the center of mass

[0031]

[0032]

[0033]

[0034] Where x, y, and z are the positions of the aircraft in the ground coordinate system, and ψ is the yaw angle of the aircraft.

[0035] D. Kinematic equations of rotation about the center of mass

[0036] From the formation process of the fuselage axis system, the projection of the aircraft's rotational angular velocity on the fuselage axis system can be written as:

[0037]

[0038]

[0039]

[0040] Solving the equation, we can get the kinematic equation of the aircraft around the center of mass:

[0041]

[0042]

[0043]

[0044] Furthermore, the specific process of step (2) is as follows:

[0045] According to the singular perturbation theory, the 12 state variables of the aircraft nonlinear dynamic model are divided into the following time scales:

[0046] ① Fast state: roll angular velocity p, pitch angular velocity q, yaw angular velocity r;

[0047] ② Slow state: angle of attack α, sideslip angle β, track roll angle μ;

[0048] ③ Very slow state: speed V, track inclination angle γ, track yaw angle χ;

[0049] ④Slowest state: x, y, z.

[0050] Furthermore, the specific process of step (3) is as follows:

[0051] Ignoring the additional forces and torques caused by the vehicle deformation, the differential equation of the angular velocity loop is expressed as the following nonlinear system:

[0052]

[0053] for The expression of the part not related to the control surface is, is the matrix for controlling the rudder surface; is a vector of 8 aircraft states, defined as: is the control vector composed of the three control surface deflection angles, defined as

[0054]

[0055] It is the output of the controller and the input of the aircraft object.

[0056] Assume that the ideal closed-loop dynamics of the angular velocity loop is:

[0057]

[0058]

[0059]

[0060] Where: p c ,q c ,r c are the command signals generated by the attitude control system. They are also the steady-state values of the fast state and will serve as the input of the slow state of the aircraft to generate the desired angular rate signal in the aircraft dynamics. is the angular velocity loop bandwidth.

[0061] From this we can calculate the desired angular acceleration Control Input Should have the form:

[0062]

[0063] Where: is a 3×3 matrix, since its right inverse It exists, and we only need to find a set of suitable steering deflection combination inputs to make the loop produce the desired angular acceleration. The calculated steering deflection [δ a δ e δ r ] T This is the output of the angular velocity control system.

[0064] Furthermore, the specific process of step (4) is as follows:

[0065] Ignoring the additional forces and moments caused by the aircraft deformation and the direct forces generated by the control surface deflection, the aircraft equation corresponding to the attitude angle loop can be written as:

[0066]

[0067] Where: for The expression of the part that is not related to p,q,r, is the coefficient matrix related to p,q,r. In order to make the output state of the aircraft track the expected value of the attitude angle α well c ,β c ,μ c , In the attitude control system, their expected values will be Instead, the expected dynamic has the following form:

[0068]

[0069]

[0070]

[0071] is the three-channel bandwidth of the attitude angle loop.

[0072] definition in [pqr] T =[p c q c r c ] T , From this, the output of the attitude control system can be solved as:

[0073]

[0074] Furthermore, the specific process of step (5) is as follows:

[0075] Ignoring the additional forces and moments caused by the aircraft deformation, we can first directly solve μ c ,get:

[0076]

[0077] is the expected dynamics of the track yaw acceleration;

[0078] Then in μ c If it is known, we can get the value of only variable α c The nonlinear equation is:

[0079]

[0080] The angle of attack command α can be obtained by applying the Newton iteration method to solve nonlinear equations c Finally, the thrust command T can be calculated c :

[0081]

[0082] in, are respectively the aircraft acceleration, the aircraft track yaw acceleration and the track pitch acceleration, Expected value in the motor generator The linear feedback and L1 adaptive control methods are used to obtain:

[0083] Taking the velocity channel as an example, the velocity tracking error e is V (t)=V(t)-V r (t), where V r (t) is the speed command signal, and the speed error vector is defined as: According to the above definition, we can further obtain the following error dynamic system equation:

[0084]

[0085] in: are the system matrix and input matrix respectively.

[0086] The control input of the error system is:

[0087] The outer loop control command signal of the speed channel can be further obtained:

[0088]

[0089] The feedback gain k is solved using the LQ (linear quadratic) method. V , the following control law of the speed channel of the main control system can be obtained:

[0090]

[0091] join in In the case of auxiliary control systems, the control input consists of two parts, namely:

[0092] u V (t) = u V,b (t)+u V,a (t)

[0093] That is, the main control system control input u V,b (t) and the control input u of the auxiliary control system V,a (t). Substitute into the control law u of the main control system V,b (t), after introducing the uncertainty of input gain and input disturbance, the following error dynamic system can be obtained:

[0094]

[0095] Among them A m,V =A V -bV k V , is an unknown real number representing the input gain; is the time-varying parameter vector, represents the input disturbance associated with the system state; represents the external input disturbance.

[0096] According to the above error dynamics system, we can get The control law of the auxiliary control system is as follows:

[0097]

[0098] Among them, r V (s) with are the reference signals r V The Laplace transform of (t), and k gv 、 is the feedback gain, D V (s) is the strictly proper transfer function, Respectively express estimates;

[0099] Establish Adaptive law:

[0100]

[0101]

[0102]

[0103] represents the estimation error of the system state, is the adaptive law, P V is the algebraic Lyapunov equation The solution, Q V is an arbitrary symmetric positive definite matrix, and Proj is the projection operator.

[0104] For the track angle control channel, including the track tilt angle channel and the track yaw angle channel, the tracking error of the track angle and the corresponding error value vector can be defined similarly. Then the error dynamic system of the track angle can also be derived, and the control signal of the track angle channel can be further obtained.

[0105] The beneficial effects brought about by adopting the above technical solution are:

[0106] (1) The present invention adopts a nonlinear dynamic inverse control method based on the six-degree-of-freedom dynamic model of the morphing aircraft, without neglecting any high-order nonlinear terms in the model, thus ensuring the control accuracy;

[0107] (2) The present invention adopts a method combining linear feedback dynamic inversion and L1 adaptation in the track control system, which can better suppress the influence of system uncertainty and input disturbance, and can maintain the accuracy of tracking control and the robustness of the system under the condition of parameter uncertainty.

[0108] (3) The present invention can be easily transplanted into the control system design of other variant aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0109] Figure 1 Schematic diagram of the L1 adaptive dynamic inverse control scheme of the variant aircraft of the present invention;

[0110] Figure 2 This is a structural block diagram of the angular velocity control system of the variant aircraft of the present invention;

[0111] Figure 3 This is a structural block diagram of the attitude control system of the variant aircraft of the present invention;

[0112] Figure 4 This is a structural block diagram of the trajectory control system of a variant aircraft of the present invention;

[0113] Figure 5 The tracking errors of the variant aircraft in the simulation example when there is an input disturbance; (a) is the tracking error of the velocity when there is an input disturbance, (b) is the tracking error of the track pitch angle when there is an input disturbance, and (c) is the tracking error of the track yaw angle when there is an input disturbance;

[0114] Figure 6 The tracking errors of the variant aircraft in the simulation example when the shape changes; (a) is the tracking error of the velocity when the shape changes, (b) is the tracking error of the track pitch angle when the shape changes, and (c) is the tracking error of the track yaw angle when the shape changes;

[0115] Figure 7 are the tracking errors of the variant aircraft in the simulation example when the input disturbance and the shape change exist at the same time. (a) is the tracking error of the velocity when the input disturbance and the shape change exist at the same time, (b) is the tracking error of the track pitch angle when the input disturbance and the shape change exist at the same time, and (c) is the tracking error of the track yaw angle when the input disturbance and the shape change exist at the same time. DETAILED DESCRIPTION

[0116] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings.

[0117] The present invention designs a flight control system based on linear feedback nonlinear dynamic inversion and L1 adaptive method. In order to achieve the above technical objectives, the technical solution of the present invention includes the following steps:

[0118] Step 1: Establish a nonlinear dynamic model of the morphing aircraft, which can characterize the dynamic characteristics of the aircraft during the morphing process;

[0119] Step 2: Based on the singular perturbation theory, the 12 state variables of the aircraft nonlinear dynamic model are time-scale separated;

[0120] Step 3: Based on the differential equation of the attitude angle loop of the morphing aircraft, the attitude angle loop control law is designed using the nonlinear dynamic inverse method;

[0121] Step 4: Based on the differential equation of the attitude angle loop of the morphing aircraft, design the attitude angle loop control law using the nonlinear dynamic inverse method;

[0122] Step 5: Based on the differential equations of the maneuvering state of the morphing aircraft, the maneuver generator control law is designed using the linear feedback nonlinear dynamic inversion and L1 adaptive method;

[0123] The flight control scheme of the variant aircraft of the present invention is as follows Figure 1 shown.

[0124] In this embodiment, the above step 1 is implemented using the following preferred solution:

[0125] A. Dynamic equations for the movement of the center of mass of an aircraft

[0126]

[0127]

[0128]

[0129] in, are the flight acceleration, angle of attack acceleration and sideslip acceleration of the aircraft respectively, V, α, β are the flight speed, angle of attack and sideslip angle of the aircraft respectively, m is the mass of the aircraft, p, q, r are the roll, pitch and yaw angular velocities of the aircraft respectively, F x ,F y ,F z are the components of the total external force acting on the aircraft in the body axis system, F ix ,F iy ,F iz are the components of the additional force generated by the aircraft due to the change in the center of mass position under the body axis system, and the calculation formula is:

[0130]

[0131] Among them, T is the engine thrust, D, Y, L are the drag, lift and side force of the aircraft respectively, g is the acceleration due to gravity, are the roll angle and pitch angle of the aircraft respectively, They are roll angular acceleration, pitch angular acceleration and yaw angular acceleration respectively; S x ,S y ,S z represents the static moment under the aircraft body axis, are the first-order derivative of the static moment with time, are the second derivatives of the static moment with time.

[0132] B. Dynamic equations for rotation around the center of mass

[0133] According to the theory of multi-rigid body dynamics, unlike conventional aircraft, a morphing aircraft will generate additional forces and torques due to deformation. The dynamic equation of the morphing aircraft rotating around its center of mass is deduced as follows:

[0134]

[0135] in, is the moment of inertia matrix of the aircraft, I x , I y , I z is the moment of inertia of the aircraft, I xy , I xz , I yz is the product of inertia. They are rolling moment, pitching moment and yaw moment respectively. They represent the components of flight acceleration on the body axis, and the expressions of b1, b2, and b3 are:

[0136]

[0137]

[0138]

[0139] Among them, u, v, w represent the components of the flight speed on the body axis, m i represents the weight of the i-th wing of the i-th aircraft, is the first derivative of the aircraft's moment of inertia with time, is the first derivative of the product of inertia with time, S ix ,S iy ,S iz Represent the static moment of the i-th wing of the aircraft, They represent the first-order derivative of the static moment of the i-th wing of the aircraft with time, They represent the second derivative of the static moment of the i-th wing of the aircraft with time.

[0140] C. Kinematic equations for the movement of the center of mass

[0141]

[0142]

[0143]

[0144] Where x, y, and z are the positions of the aircraft in the ground coordinate system, and ψ is the yaw angle of the aircraft.

[0145] D. Kinematic equations of rotation about the center of mass

[0146] From the formation process of the fuselage axis system, the projection of the aircraft's rotational angular velocity on the fuselage axis system can be written as:

[0147]

[0148]

[0149]

[0150] Solving the equation, we can get the kinematic equation of the aircraft around the center of mass:

[0151]

[0152]

[0153]

[0154] The forces and moments acting on the morphing vehicle are calculated as follows:

[0155] Aerodynamic forces include side force Y, drag D and lift L, and aerodynamic moments include rolling moment Pitching moment and yaw moment

[0156] The aerodynamic force and torque module uses feedback data from the flight simulation system (such as Mach number, altitude, angle of attack, sideslip angle, aircraft center of gravity position and angular velocity and other flight parameters and control surface positions, and the positions of landing gear and flaps provided by the hydraulic system) to calculate the aerodynamic coefficients on the stabilizing axis. Finally, the aerodynamic force and torque on the fuselage axis are calculated and output to the aircraft six-degree-of-freedom motion model module. The aerodynamic force is mainly calculated based on the aerodynamic pressure and the aerodynamic force coefficient on the stabilizing axis. The calculation formulas for the aerodynamic force coefficients are:

[0157]

[0158] The calculation formulas for lift, drag and side force are:

[0159]

[0160] The aerodynamic moment is mainly calculated based on the aerodynamic pressure and the aerodynamic moment coefficient on the stabilizing axis. The calculation formulas for the pitch moment coefficient, rolling moment coefficient, and yaw moment coefficient are as follows:

[0161]

[0162] The calculation formulas for pitching moment, rolling moment, and yaw moment are:

[0163]

[0164] Where ρ is the air density, S is the wing area, b is the wing span, and c is the average aerodynamic chord length of the wing. L 、C D 、C Y 、C m 、C l 、C n are the lift coefficient, drag coefficient, side force coefficient, pitch moment coefficient, roll moment coefficient and yaw moment coefficient of the morphing aircraft respectively, η is the measure of the deformation of the morphing aircraft, are the corresponding aerodynamic coefficients that vary with the aircraft's shape. Next, the morphing aircraft is dynamically modeled. Based on the aerodynamic derivatives obtained above, a six-degree-of-freedom nonlinear multibody dynamic model for the morphing aircraft is derived.

[0165] In this embodiment, the above step 2 is implemented using the following preferred solution:

[0166] Since the six-degree-of-freedom nonlinear dynamic model of the variant aircraft is a dynamic system with 12 state variables, the 12 state variables are velocity V, angle of attack α, sideslip angle β, roll angular velocity p, pitch angular velocity q, yaw angular velocity r, track roll angle μ, track tilt angle γ, track yaw angle χ, and the coordinates x, y, and z of the center of mass position projected on the horizontal plane. The response speed of these 12 states to the control instructions is different. If the above nonlinear dynamic inverse method is to be used to complete the input / output linearization of the speed control channel and the track angle (including track deviation angle and track tilt angle) control channel, the aircraft's six-degree-of-freedom nonlinear dynamic equations need to be divided into four subsystems according to the different response speeds of the state variables. According to the singular perturbation theory, the time scale can be divided into:

[0167] ① Fast state: roll angular velocity p, pitch angular velocity q, yaw angular velocity r;

[0168] ② Slow state: angle of attack α, sideslip angle β, track roll angle μ;

[0169] ③ Very slow state: speed V, track inclination angle γ, track yaw angle χ;

[0170] ④Slowest state: x, y, z.

[0171] These four states—fast, slow, very slow, and slowest—form four dynamic subsystems: the angular velocity control system, attitude control system, track control system, and position control system, which in turn form the larger flight control system. Within the larger system, different bandwidths are selected for each subsystem based on the principle of time-scale separation, ensuring that each dynamic subsystem operates in a distinct time domain. For each subsystem, the nonlinear dynamic inverse calculation method described above is used to perform input / output linearization, ultimately achieving linearization of the velocity control channel and the track angle control channel.

[0172] In this embodiment, the above step 3 is implemented using the following preferred solution:

[0173] Ignoring the additional forces and torques caused by the vehicle deformation, the differential equation of the angular velocity loop is expressed as the following nonlinear system:

[0174]

[0175] for The expression of the part not related to the control surface is, is the matrix for controlling the rudder surface; is a vector of 8 aircraft states, defined as: is the control vector composed of the three control surface deflection angles, defined as

[0176]

[0177] It is the output of the controller and the input of the aircraft object.

[0178] Assume that the ideal closed-loop dynamics of the angular velocity loop is:

[0179]

[0180]

[0181]

[0182] Where: p c ,q c ,r c are the command signals generated by the attitude control system. They are also the steady-state values of the fast state and will serve as the input of the slow state of the aircraft to generate the desired angular rate signal in the aircraft dynamics. is the angular velocity loop bandwidth, which is taken as 10 rad / s.

[0183] From this we can calculate the desired angular acceleration Control Input Should have the form:

[0184]

[0185] Where: is a 3×3 matrix, since its right inverse It exists, and we only need to find a set of suitable steering deflection combination inputs to make the loop produce the desired angular acceleration. The calculated steering deflection [δ a δ e δ r ] T This is the output of the angular velocity control system.

[0186] The structural block diagram of the angular velocity control system is as follows Figure 2 shown.

[0187] In this embodiment, the above step 4 is implemented using the following preferred solution:

[0188] Ignoring the additional forces and moments caused by the aircraft deformation and the direct forces generated by the control surface deflection, the aircraft equation corresponding to the attitude angle loop can be written as:

[0189]

[0190] Where: for The expression of the part that is not related to p,q,r, is the coefficient matrix related to p,q,r. In order to make the output state of the aircraft track the expected value of the attitude angle α well c ,β c ,μ c , In the attitude control system, their expected values will be Instead, the expected dynamic has the following form:

[0191]

[0192]

[0193]

[0194] is the three-channel bandwidth of the attitude angle loop.

[0195] definition in [pqr] T =[p c q c r c ] T , From this, the output of the attitude control system can be solved as:

[0196]

[0197] The structural diagram of the attitude control system is as follows: Figure 3 shown.

[0198] In this embodiment, the above step 5 is implemented using the following preferred solution:

[0199] Ignoring the additional forces and moments caused by the aircraft deformation, we can first directly solve μ c ,get:

[0200]

[0201] is the expected dynamics of the track yaw acceleration;

[0202] Then in μ c If it is known, we can get the value of only variable α c The nonlinear equation is:

[0203]

[0204] The angle of attack command α can be obtained by applying the Newton iteration method to solve nonlinear equations c Finally, the thrust command T can be calculated c :

[0205]

[0206] in, are respectively the aircraft acceleration, the aircraft track yaw acceleration and the track pitch acceleration, Expected value in the motor generator The linear feedback and L1 adaptive control methods are used to obtain:

[0207] Taking the velocity channel as an example, the velocity tracking error e is V (t)=V(t)-V r (t), where V r (t) is the speed command signal, and the speed error vector is defined as: According to the above definition, we can further obtain the following error dynamic system equation:

[0208]

[0209] in:

[0210]

[0211] are the system matrix and input matrix respectively.

[0212] The control input of the error system is:

[0213]

[0214] The outer loop control command signal of the speed channel can be further obtained:

[0215]

[0216] The feedback gain k is solved using the LQ (linear quadratic) method. V , the following control law of the speed channel of the main control system can be obtained:

[0217]

[0218] join in In the case of auxiliary control systems, the control input consists of two parts, namely:

[0219] u V (t) = u V,b (t)+u V,a (t)

[0220] That is, the main control system control input u V,b (t) and the control input u of the auxiliary control system V,a (t). Substitute into the control law u of the main control system V,b (t), and after introducing the uncertainty of the input gain and the input disturbance, the following error dynamics system can be obtained:

[0221]

[0222] Among them A m,V =A V -b V k V , is an unknown real number representing the input gain; is the time-varying parameter vector, represents the input disturbance associated with the system state; represents the external input disturbance.

[0223] The goal of L1 adaptive control design is to ensure that the system output effectively tracks the given command signal. The design is based on the following three assumptions:

[0224] a. Unknown parameter θ V(t) and σ V (t) uniformly bounded;

[0225] b. The rate of change of the parameter over time is uniformly bounded;

[0226] c. The upper and lower bounds of the uncertain input gain are known. It should be noted that as a representation of the input disturbance of the real physical system, θ V (t) and σ V The boundedness of (t) and the corresponding time rate of change is naturally guaranteed; for the input gain, although the exact upper and lower bounds are not available, larger limits can be set in the design to improve the system's ability to cope with gain uncertainty.

[0227] For the error dynamics system proposed above, consider the following state estimator:

[0228]

[0229] Among them, adaptive estimation and It is obtained by the following adaptive law:

[0230]

[0231]

[0232]

[0233] in, represents the estimation error of the system state, is the adaptive law, P V is the algebraic Lyapunov equation The solution (Q V is an arbitrary symmetric positive definite matrix), Proj is the projection operator.

[0234] According to the above error dynamics system, we can get The control law of the auxiliary control system is as follows:

[0235]

[0236] Among them, r V (s) with are the reference signals r V The Laplace transform of (t) and k gv 、 is the feedback gain, D V (s) is a strictly proper transfer function, and from this we can obtain the following strictly proper and stable transfer function:

[0237]

[0238] And there is C V (0)=1, considering that the design goal of the control system in this problem is to track the speed command, so we take r V (t) = 0, and the control law can be simplified as follows:

[0239]

[0240] For the track angle control channel (including the track tilt angle channel and the track yaw angle channel), the tracking error of the track angle and the corresponding error value vector can be defined similarly. Then the error dynamic system of the track angle can also be derived, and the control signal of the track angle channel can be further obtained.

[0241] The structural block diagram of the track control system is as follows Figure 4 .

[0242] The simulation scenario is set as follows: the initial condition of the simulation is the trim cruise state (h = 5000m, V = 200m / s), followed by command signal tracking. The command tracking signal is described as follows:

[0243] At t=0s, a speed step instruction of +10m / s is given, and a reference trajectory is generated through the filter as the target to be tracked by the variant aircraft model; at t=20s, a track tilt angle step instruction of +0.15rad is given, and a reference trajectory is generated through the filter as the target to be tracked by the variant aircraft model; at t=40s, a track yaw angle step instruction of +0.2rad is given, and a reference trajectory is generated through the filter as the target to be tracked by the variant aircraft model.

[0244] Simulations are performed in the presence of input disturbances and changes in aircraft shape. The simulation experiment conditions and numbers are shown in the following table:

[0245] Table: Simulation test condition settings

[0246]

[0247] Test Case I will study the impact of input disturbances on the control performance of the aircraft. The added input disturbance is generated by a signal with a reference value of 0 and a continuous normal distribution. The disturbance is directly superimposed on the control input signal generated by the controller; the disturbance signal superimposed on the throttle push rod value and each aerodynamic control surface fluctuates randomly within 10% of the original control signal, that is, σPLA ％ ∈[-0.1,0.1],

[0248] Test scenario II will simulate the variation process of the variant aircraft. During the simulation process, the variation process of the aircraft is simulated by changing the aircraft span, average aerodynamic chord length, moment of inertia and related aerodynamic parameters (the range and variation rules of the aircraft mass parameters and aerodynamic parameters are obtained through relevant literature data).

[0249] Test case III will combine test cases I and II to test the control effect of the control system under the simultaneous existence of input disturbances and parameter uncertainties.

[0250] Depend on Figure 5-Figure 7 As can be seen, after adding the L1 adaptive dynamic inverse control system, the aircraft model can still effectively track the reference trajectory of the speed and track angle command signals during the deformation process. The tracking errors of each output quantity can be stably converged. The errors are within the engineering allowable range throughout the simulation process, and the system remains in a stable operating state. The simulation results show that the L1 adaptive dynamic inverse controller can enable the flight control system to have high tracking control accuracy and the ability to cope with changes in flight parameters, thereby ensuring the stability of the morphing aircraft during the deformation process.

[0251] The embodiments are only for illustrating the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention fall within the protection scope of the present invention.

Claims

1. A method for controlling a morphing aircraft based on L1 adaptive dynamic inversion, characterized in that: The following steps are involved: (1) Establish a nonlinear dynamic model of the morphing aircraft, which can characterize the dynamic characteristics of the aircraft during the morphing process; (2) Based on the singular perturbation theory, the 12 state variables of the aircraft nonlinear dynamics model are time-scale separated; (3) According to the differential equation of the angular velocity loop of the morphing aircraft, the angular velocity loop control law is designed based on the nonlinear dynamic inverse method; (4) According to the differential equation of the attitude angle loop of the morphing aircraft, the attitude angle loop control law is designed based on the nonlinear dynamic inverse method; (5) Based on the differential equations of the maneuvering state of the morphing aircraft, the maneuver generator control law is designed using the linear feedback nonlinear dynamic inversion and L1 adaptive method; In step (1), the nonlinear dynamic model of the variant aircraft described includes: A. Dynamic equations for the movement of the center of mass of an aircraft in, are the flight acceleration, angle of attack acceleration and sideslip acceleration of the aircraft respectively, V, α, β are the flight speed, angle of attack and sideslip angle of the aircraft respectively, m is the mass of the aircraft, p, q, r are the roll, pitch and yaw angular velocities of the aircraft respectively, F x ,F y ,F z are the components of the total external force acting on the aircraft in the body axis system, F ix ,F iy ,F iz are the components of the additional force generated by the aircraft due to the change in the center of mass position under the body axis system, and the calculation formula is: Among them, T is the engine thrust, D, Y, L are the drag, lift and side force of the aircraft respectively, g is the acceleration of gravity, φ, θ are the roll angle and pitch angle of the aircraft respectively, They are roll angular acceleration, pitch angular acceleration and yaw angular acceleration respectively; S x ,S y ,S z represents the static moment under the aircraft body axis, are the first-order derivative of the static moment with time, are the second derivative of the static moment with time; B. Dynamic equations for rotation around the center of mass According to the theory of multi-rigid body dynamics, unlike conventional aircraft, a morphing aircraft will generate additional forces and torques due to deformation. The dynamic equation of the morphing aircraft rotating around its center of mass is deduced as follows: in, is the moment of inertia matrix of the aircraft, I x , I y , I z is the moment of inertia of the aircraft, I xy , I xz , I yz is the product of inertia; M and N are rolling moment, pitching moment and yaw moment respectively; They represent the components of flight acceleration on the body axis, and the expressions of b1, b2, and b3 are: Among them, u, v, w represent the components of the flight speed on the body axis, m i represents the weight of the i-th wing of the i-th aircraft, is the first-order derivative of the aircraft's moment of inertia with time, is the first derivative of the product of inertia with time, S ix ,S iy ,S iz Represent the static moment of the i-th wing of the aircraft, They represent the first-order derivative of the static moment of the i-th wing of the aircraft with time, They represent the second derivative of the static moment of the i-th wing of the aircraft with time; C. Kinematic equations for the movement of the center of mass Where x, y, z are the positions of the aircraft in the ground coordinate system, and ψ is the yaw angle of the aircraft; D. Kinematic equations of rotation about the center of mass From the formation process of the fuselage axis system, the projection of the aircraft's rotational angular velocity on the fuselage axis system can be written as: Solving the equation, we can get the kinematic equation of the aircraft around the center of mass:

2. The method for controlling a morphing aircraft based on L1 adaptive dynamic inversion according to claim 1, characterized in that: The specific process of step (2) is as follows: According to the singular perturbation theory, the 12 state variables of the aircraft nonlinear dynamic model are divided into the following time scales: ① Fast state: roll angular velocity p, pitch angular velocity q, yaw angular velocity r; ② Slow state: angle of attack α, sideslip angle β, track roll angle μ; ③ Very slow state: speed V, track inclination angle γ, track yaw angle χ; ④Slowest state: x, y, z.

3. The method for controlling a morphing aircraft based on L1 adaptive dynamic inversion according to claim 1, characterized in that: The specific process of step (3) is as follows: Ignoring the additional forces and torques caused by the vehicle deformation, the differential equation of the angular velocity loop is expressed as the following nonlinear system: for The expression of the part not related to the control surface is, is the matrix for controlling the rudder surface; is a vector of 8 aircraft states, defined as: is the control vector composed of the three control surface deflection angles, defined as It is the output of the controller and also the input of the aircraft object; Assume that the ideal closed-loop dynamics of the angular velocity loop is: Where: p c ,q c ,r c are the command signals generated by the attitude control system. They are also the steady-state values of the fast state and will serve as the input of the slow state of the aircraft to generate the desired angular rate signal in the aircraft dynamics. is the angular velocity loop bandwidth; From this we can calculate the desired angular acceleration Control Input Should have the form: Where: is a 3×3 matrix, since its right inverse It exists, and we only need to find a set of suitable steering deflection combination inputs to make the loop produce the desired angular acceleration; the calculated steering deflection [δ a δ e δ r ] T This is the output of the angular velocity control system.

4. The method for controlling a morphing aircraft based on L1 adaptive dynamic inversion according to claim 1, characterized in that: The specific process of step (4) is as follows: Ignoring the additional forces and moments caused by the aircraft deformation and the direct forces generated by the control surface deflection, the aircraft equation corresponding to the attitude angle loop can be written as: Where: for The expression of the part that is not related to p,q,r, is the coefficient matrix related to p, q, and r; in order to make the output state of the aircraft track the expected value of the attitude angle α well c ,β c ,μ c , In the attitude control system, their expected values will be Instead, the expected dynamic has the following form: is the three-channel bandwidth of the attitude angle loop; definition medium[pqr] T =[p c q c r c ] T , From this, the output of the attitude control system can be solved as:

5. The method for controlling a morphing aircraft based on L1 adaptive dynamic inversion according to claim 1, characterized in that: The specific process of step (5) is as follows: Ignoring the additional forces and moments caused by the aircraft deformation, we can first directly solve μ c ,get: is the expected dynamics of the track yaw acceleration; Then in μ c If it is known, we can get the value of only variable α c The nonlinear equation is: The angle of attack command α can be obtained by applying the Newton iteration method to solve nonlinear equations c Finally, the thrust command T can be calculated c : in, are respectively the aircraft acceleration, the aircraft track yaw acceleration and the track pitch acceleration, Expected value in the motor generator The linear feedback and L1 adaptive control methods are used to obtain: Let velocity tracking error e V (t)=V(t)-V r (t), where V r (t) is the speed command signal, and the speed error vector is defined as: According to the above definition, we can further obtain the following error dynamic system equation: in: are the system matrix and input matrix respectively; The control input of the error system is: The outer loop control command signal of the speed channel can be further obtained: The feedback gain k is solved using the LQ method. V , the following control law of the speed channel of the main control system can be obtained: join in In the case of auxiliary control systems, the control input consists of two parts, namely: u V (t)=u V,b (t)+u V,a (t) That is, the main control system control input u V,b (t) and the control input u of the auxiliary control system V,a (t); Substitute into the control law u of the main control system V,b (t), after introducing the uncertainty of input gain and input disturbance, the following error dynamic system can be obtained: Among them A m,V =A V -b V k V , is an unknown real number representing the input gain; is the time-varying parameter vector, represents the input disturbance associated with the system state; represents the external input disturbance; According to the above error dynamics system, we can get The control law of the auxiliary control system is as follows: Among them, r V (s) with are the reference signals r V The Laplace transform of (t), and is the feedback gain, D V (s) is the strictly proper transfer function, Respectively represent ω V , σ V (t) estimate; Establish Adaptive law: represents the estimation error of the system state, is the adaptive law, P V is the algebraic Lyapunov equation The solution, Q V is an arbitrary symmetric positive definite matrix, Proj is the projection operator; For the track angle control channel, including the track tilt angle channel and the track yaw angle channel, the tracking error of the track angle and the corresponding error value vector can be defined similarly. Then the error dynamic system of the track angle can also be derived, and the control signal of the track angle channel can be further obtained.

6. The method for controlling a morphing aircraft based on L1 adaptive dynamic inversion according to claim 1, characterized in that: The forces and moments acting on the morphing vehicle are calculated as follows: Aerodynamic forces include side force Y, drag D and lift L, and aerodynamic moments include rolling moment Pitching moment M and yaw moment N; The aerodynamic force and torque module uses feedback data from the flight simulation system to calculate the aerodynamic coefficients on the stabilizing axis. Finally, it calculates the aerodynamic force and torque on the body axis and outputs them to the aircraft six-degree-of-freedom motion model module. The aerodynamic force is mainly calculated based on the aerodynamic pressure and the aerodynamic force coefficient on the stabilizing axis. The calculation formulas for the aerodynamic force coefficient are: The calculation formulas for lift, drag and side force are: The aerodynamic moment is mainly calculated based on the aerodynamic pressure and the aerodynamic moment coefficient on the stabilizing axis. The calculation formulas for the pitch moment coefficient, rolling moment coefficient, and yaw moment coefficient are: The calculation formulas for pitching moment, rolling moment, and yaw moment are: Where ρ is the air density, S is the wing area, b is the wing span, and c is the average aerodynamic chord length of the wing; C L 、C D 、C Y 、C m 、C l 、C n are the lift coefficient, drag coefficient, side force coefficient, pitch moment coefficient, roll moment coefficient and yaw moment coefficient of the morphing aircraft respectively, η is the measure of the deformation of the morphing aircraft, C L 、C D 、C Y 、C m 、C l 、C n The corresponding aerodynamic coefficients in the expression change with the shape of the aircraft.

Images