Dual-channel coupled control method, system and control device for supersonic tailless aircraft

Abstract

The application discloses a supersonic tailless aircraft double-channel coupling control method, system and control device, and belongs to the technical field of aircraft control. The control method firstly establishes an attitude control model of the supersonic tailless aircraft, then mathematically describes safety constraints and performance constraints of attack angle and sideslip angle coupling of the supersonic tailless aircraft, and defines 'control constraints'. Then, aiming at the channel coupling and constraint coupling problem, two processing ideas of coupling instruction filter and coupling funnel control are respectively proposed. Finally, based on the nonlinear dynamic inverse design idea, the coupling funnel controller is designed. The control method in the application is verified by simulation experiments, and not only can meet the safety constraints and performance constraints of STAV attitude control, but also has strong robustness and can be applied to two typical task scenes of high-altitude penetration and low-altitude penetration.

Description

Technical Field

[0001] This invention relates to the field of aircraft control technology, and in particular to a dual-channel coupled control method, system and control device for supersonic tailless aircraft. Background Technology

[0002] Penetrating air superiority, aimed at breaking through enemy defense zones and conducting effective reconnaissance, deep strikes, and airspace clearance to gain air supremacy, is a crucial operational style in future air warfare. The all-aspect, extremely low stealth capability afforded by the supersonic tailless aerial vehicle (STAV) configuration allows it to penetrate high-intensity anti-access / area denial (A2 / AD) environments, making it the optimal aerodynamic design choice for next-generation fighters. However, the lack of traditional horizontal and vertical stabilizers in STAVs makes pitch and yaw control more challenging. To ensure proper engine intake operation and efficient tail-mounted aerodynamic control, limits must be placed on the angle of attack and sideslip angle ranges—the so-called safety constraints. Furthermore, to achieve effective attitude control or high-precision trajectory tracking, limits must be placed on the tracking error of the airflow angle—the so-called performance constraints. Therefore, research on STAV attitude tracking control methods for penetrating air superiority missions faces the following problems and challenges: First, during high-altitude penetration, the angle of attack and sideslip angle of the STAV must be maintained within a suitable range to ensure the normal operation of the air intake, especially ensuring that the engine combustion efficiency is at its optimal state during the cruise phase; second, during low-altitude penetration, excessive angles of attack and sideslip angles must be avoided, as this can generate excessive turbulence at the front of the aircraft, causing the aerodynamic control surfaces at the rear of the aircraft to be completely immersed in the low-energy stall wake, resulting in a sharp decline in control surface control effectiveness; third, the airflow angle needs to maintain extremely high tracking accuracy in order to further achieve high-precision trajectory tracking to penetrate the defense zone or avoid ground obstacles. Therefore, research on the performance constraints of STAV dual-channel coupling is of great significance.

[0003] In existing control theory research, state constraints are typically implemented using barrier Lyapunov functions, while error constraints are usually achieved through preset performance control or funnel control. All three control methods are essentially nonlinearly tuned proportional controls. Funnel control was the earliest proposed concept and provides a more vivid description of the control process. Funnel control refers to designing a control law that allows the tracking error to converge to an adjustable set of residuals with a preset convergence rate and maximum overshoot. Typical issues addressed in theoretical research include: unknown control direction, and how to handle input nonlinearity and unknown nonlinear terms. Bechlioulis CP et al. transformed a class of non-affine pure feedback systems into pseudo-affine systems based on the Lagrange mean value theorem and designed a funnel controller that requires no estimator. For strict feedback systems with dead-zone input nonlinearity, Theodorakopoulos A et al. designed a funnel controller using a backpropagation method and extended this method to multi-input multi-output linear systems.

[0004] In the field of flight control research, most existing research focuses on extending the funnel control method proposed by scholars such as Ilchmann A and Bechlioulis to other more complex systems. However, there is limited improvement on the funnel control method itself, and few studies apply it to STAV flight control. In practical engineering, both STAV attitude angle control and trajectory tracking control aim to achieve non-rectangular domain constraints that conform to the characteristics of the aircraft. However, most existing research constructs boundary functions separately for the decoupled channels, resulting in controllers that cannot achieve dual-channel coupling performance constraints or simultaneously address safety and error constraints. Since the funnel control method can only constrain attitude angle tracking errors, further research is needed on how to achieve non-rectangular domain state constraints by combining command signals and tracking errors, and how to ensure actuator constraints while satisfying STAV attitude angle constraints. Summary of the Invention

[0005] To address the aforementioned problems, this invention aims to provide a dual-channel coupled control method, system, and control device for a supersonic tailless vehicle (STAV). The method first mathematically describes the safety and performance constraints of the STAV's angle of attack and sideslip angle coupling and defines "control constraints." Then, for the channel coupling and constraint coupling problems, two processing approaches are proposed: a coupled command filter and a coupled funnel control. Finally, based on the concept of nonlinear dynamic inverse design, a coupled funnel controller is designed. This control method not only satisfies the safety and performance constraints of STAV attitude control but also exhibits strong robustness.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A dual-channel coupled control method for a supersonic tailless vehicle includes the following steps.

[0008] S1: Establish an attitude control model for a supersonic tailless aircraft;

[0009] S2: Based on dual-channel coupling constraints, determine the control constraint conditions for the coupled control of a supersonic tailless vehicle;

[0010] S3: Design a coupled funnel controller to control the attitude and torque distribution of a supersonic tailless aircraft.

[0011] Furthermore, the attitude control model of the supersonic tailless aircraft described in step S1 is as follows:

[0012]

[0013] In the formula, For the STAV system status, For aerodynamic torque, This is the aerodynamic torque coefficient; This is the transformation matrix from the body coordinate system to the airflow coordinate system. This is the transformation matrix from torque to attitude angle and angular rate. This is the transformation matrix from torque coefficient to torque; Indicates dynamic pressure. Indicates the reference area of ​​the aircraft; The vector composed of aerodynamic control surfaces; This is due to unknown external interference. It is a nonlinear function related to the system state; It is a non-affine function.

[0014] Furthermore, the specific operation of step S2 includes the following steps:

[0015] S201: Determine the angle of attack of a supersonic tailless vehicle With sideslip angle Dual-channel non-rectangular domain security constraints are proposed, and a mathematical description of the non-rectangular domain security constraints is provided.

[0016] S202: Design a dual-channel coupled command filter;

[0017] S203: Based on performance constraints and safety constraints, define dual-channel coupled control constraints and determine control constraint conditions.

[0018] Furthermore, the dual-channel coupling command filter in step S202 is...

[0019]

[0020] Wherein, subscript c represents control signal and subscript d represents command signal; Represents the hyperbolic cosine function. Represents the hyperbolic tangent function. and The compensation parameters for the filter to be designed are as follows: and Convergence parameters of the filter to be designed.

[0021] Furthermore, the specific operation of step S203 includes the following steps:

[0022] S2031: Define the airflow angle tracking error as follows: , and Then the airflow angle error satisfies the error constraint.

[0023]

[0024] In the formula, Let be the error envelope function;

[0025] S2032: Performance constraints are defined based on error constraints.

[0026]

[0027] in, , ;

[0028] S2033: Based on dual-channel coupling constraints, the security constraints are defined as follows:

[0029]

[0030] in, , ;

[0031] S2034: Based on performance constraints and security constraints, the control constraints are defined as follows:

[0032]

[0033] in, , This is a compensation factor.

[0034] Furthermore, the specific operation of step S3 includes the following steps:

[0035] S301: Rewrite the attitude control model of a supersonic tailless aircraft as a tracking error differential equation.

[0036]

[0037] in, This is an airflow angle command signal. For the virtual angular velocity command to be designed, To calculate the tracking error between control levels, The three-axis torques to be assigned, For torque distribution error, The virtual torque coefficient to be assigned;

[0038] S302: Design of a Dual-Channel Coupled Funnel Controller With the torque control distribution law, the STAV airflow angle is adjusted. Able to track stable command signals .

[0039] Furthermore, the steps described in step S302 The control law is

[0040]

[0041] The control law is

[0042]

[0043] In the formula, For virtual control input, Input torque for virtual control; These are the roll, pitch, and yaw rates, respectively. and It is a positive definite diagonal matrix. This represents the error transformation function. The proportional gain to be designed for each channel. This represents the standardized error variable.

[0044] Furthermore, the torque control distribution law described in step S302 is as follows:

[0045]

[0046] In the formula, It is the control input when the control surface is at zero deflection. This is the current control input. It is a non-affine function. It is the difference between the next time step and the current rudder deflection angle, which is the quantity to be designed; and The objective function consists of two parts, representing the torque coefficient generated by the current rudder deflection angle and the torque coefficient required to track the attitude command signal, respectively: The aim is to minimize torque distribution error; Aimed at minimizing overall rudder offset, The weight matrix, This is the incremental rudder effect matrix.

[0047] Furthermore, the present invention also includes a dual-channel coupled control system for a supersonic tailless vehicle, the control system specifically including an attitude control model establishment module, a control constraint definition module, and a control module;

[0048] The attitude control model establishment module is used to establish the attitude control model of the supersonic tailless aircraft.

[0049] The control constraint definition module is used to determine the control constraint conditions for the coupled control of a supersonic tailless vehicle.

[0050] The controller module is used to control the attitude and torque distribution of the supersonic tailless aircraft;

[0051] The attitude control model establishment module, control constraint definition module, and control module are implemented based on the control method described above.

[0052] Furthermore, the present invention also includes a dual-channel coupled control device for a supersonic tailless vehicle, the control device specifically including at least one processor; and a memory communicatively connected to the processor; wherein the memory stores instructions executable by the processor, the instructions being executed by the processor to enable the processor to perform the control method as described above.

[0053] The beneficial effects of this invention are:

[0054] 1. The dual-channel coupled control method for supersonic tailless aircraft in this invention, from an overall perspective, firstly establishes an attitude control model for the supersonic tailless aircraft, then mathematically describes the safety and performance constraints of the coupling between the angle of attack and sideslip angle of the supersonic tailless aircraft, and defines "control constraints." Next, for the channel coupling and constraint coupling problems, two processing approaches are proposed: coupled command filter and coupled funnel control. Finally, based on the nonlinear dynamic inverse design concept, a coupled funnel controller is designed. Simulation experiments verify that this control method not only meets the safety and performance constraints of STAV attitude control, but also has strong robustness and is applicable to two typical mission scenarios: high-altitude penetration and low-altitude infiltration.

[0055] 2. This invention addresses the problem of the coupling between the STAV angle of attack and sideslip angle safety constraints by proposing a non-rectangular feasible region mathematical description method dominated by the angle of attack. By dividing the non-rectangular region into several continuous trapezoidal regions that are easy to describe, the dual-channel coupled safety constraints can be accurately characterized, avoiding the complex process of polar coordinate transformation.

[0056] 3. This invention designs a coupled command filter that can compensate for tracking errors in the trajectory angle, solving the coupling problem between the upper and lower limit channels of the angle of attack and sideslip angle; in addition, by defining "control-oriented constraints", a unified description of safety constraints and error constraints is achieved, creating conditions for designing a dual-channel coupled funnel controller.

[0057] 4. This invention, based on the STAV wind tunnel data model, combines incremental control allocation and nonlinear dynamic inverse control methods to design a dual-channel coupled funnel controller that requires no estimator. The controller not only meets the safety and performance constraints of attitude control in STAV high-altitude / low-altitude penetration missions but also exhibits strong robustness. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of the control surfaces of the ICE aircraft in this invention.

[0059] Figure 2 The trapezoidal domain description method used in this invention describes the airflow angle of the aircraft. A schematic diagram illustrating the processing of the constrained region.

[0060] Figure 3 This is a schematic diagram of the performance constraint and safety constraint region processing scheme in this invention, taking the angle of attack as an example.

[0061] Figure 4 This is a flowchart illustrating the design of the STAV attitude controller based on a coupling command filter in this invention.

[0062] Figure 5 For the simulation experiment of the present invention, the aircraft Schematic diagram of constraint region processing scheme.

[0063] Figure 6 The results show the STAV attitude angle tracking performance in Simulation 1 of this invention.

[0064] Figure 7 The results of STAV attitude angle and angular rate tracking error in simulation 1 of this invention are shown.

[0065] Figure 8 The results of STAV attitude angular rate and torque in simulation 1 of this invention are shown.

[0066] Figure 9 The results of the STAV aerodynamic control surface deflection angle in simulation 1 of this invention are shown.

[0067] Figure 10 The figure shows the adaptive parameter variation curve of the controller in Case 1 of the present invention.

[0068] Figure 11The conversion error curve of the controller in Case 2 of the simulation 1 of this invention is shown.

[0069] Figure 12 The results show the STAV attitude angle tracking performance in Simulation 2 of this invention.

[0070] Figure 13 The results of STAV attitude angle and angular rate tracking error in simulation 2 of this invention are shown.

[0071] Figure 14 The results of the STAV attitude angular rate and torque in simulation 2 of this invention are shown.

[0072] Figure 15 The curve of the deflection angle of the STAV aerodynamic control surface in simulation 2 of this invention.

[0073] Figure 16 The conversion error curve of the controller in Case 4 of the simulation 2 of this invention is shown. Detailed Implementation

[0074] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0075] Example 1:

[0076] Example 1 provides a dual-channel coupled control method for a supersonic tailless vehicle, including the following steps:

[0077] Step 1: Establish an attitude control model for the supersonic tailless aircraft;

[0078] Based on kinematic and dynamic analysis, the mathematical model for the attitude control of the STAV (Supersonic Tailless Vehicle) can be expressed in the following form, which includes affine and non-affine functions:

[0079] (1)

[0080] In the formula, For the STAV system status, For aerodynamic torque, This is the aerodynamic torque coefficient; This is the transformation matrix from the body coordinate system to the airflow coordinate system. This is the transformation matrix from torque to attitude angle and angular rate. This is the transformation matrix from torque coefficient to torque; Indicates dynamic pressure. Indicates the reference area of ​​the aircraft; Vector composed of aerodynamic control surfaces ;definition ,but , and These are the deflection angles. The range of amplitude and rate of change; Indicates the deflection angle of the inner leading edge flap. Indicates the deflection angle of the outer leading edge flap. Indicates the deflection angle of the all-moving wingtip. Indicates the elliptic deflection angle. Indicates the deflection angle of the embedded surface. This indicates the pitch flap deflection angle; in the subscript, l indicates the left side and r indicates the right side. Due to unknown external interference, A nonlinear function that is dependent on the system state. A non-affine function. It can be further written in the following form

[0081]

[0082] in, These are the roll, pitch, and yaw moment coefficients, respectively, affected by the system state and the actuators.

[0083] Appendix Figure 1 This is a schematic diagram of the control surfaces of the ICE aircraft. Table 1 below lists the names, symbols, and related physical parameters of the actuators for the ICE tailless aircraft. The dynamic characteristics of the actuators are characterized by the following transfer functions.

[0084]

[0085] Table 1. Names, symbols, and related physical parameters of actuators for ICE tailless aircraft

[0086] Assumption 1: The uncertainty and external disturbances of the STAV model are bounded, and the command signal and its first and second derivatives are bounded.

[0087] Assumption 2: Under specific flight speed and altitude conditions, the STAV flight envelope is determined and known, all states of the closed-loop control system are measurable, and measurement errors and time delays are negligible.

[0088] Control objective: Based on the concept of nonlinear dynamic inverse, a dual-channel coupled funnel controller is designed to achieve the desired STAV airflow angle. Able to stably track command signals The closed-loop system has semi-globally consistent bounded signals (semi-globally consistent bounded means that the controller can only eventually converge the system state (or error) to a given sphere if the initial state (or error) is within the assumed range), and the actuator... The amplitude and bandwidth constraints are met; the angle of attack and sideslip angle of the STAV meet the non-rectangular domain safety constraints; under the condition of meeting the safety constraints, the airflow angle tracking error of the STAV is... and It can meet the pre-defined performance constraints; among which, These represent the roll angle, angle of attack, and sideslip angle in the velocity coordinate system, respectively, with the subscript d indicating an ideal signal.

[0089] Step 2: Determine the control constraints for the coupling control of the supersonic tailless aircraft;

[0090] 2.1 Analysis of STAV Dual-Channel Coupling Constraint Problem

[0091] Compared to sideslip angle and angle of attack, the roll angle has a smaller impact on STAV flight performance. Therefore, this invention mainly discusses the dual-channel coupling constraints of sideslip angle and angle of attack, aiming to lay the foundation for the design of a coupled funnel controller. Coupled funnel control refers to two aspects: first, the coupling of safety constraints between channels; and second, the coupling of safety constraints and performance constraints.

[0092] 2.2 Description of Safety Constraints for Angle of Attack and Sideslip Angle

[0093] Assuming that within a specific range of flight speed and altitude, the angle of attack of the STAV... With sideslip angle Non-rectangular domain security constraints must be met: Considering that the STAV actuator has a stronger control capability over the angle of attack compared to the sideslip angle, this invention takes the angle of attack as the main factor and uses mathematical description of the safety constraints of the non-rectangular domain. Specifically, the trapezoidal domain description method is used to describe the feasible domains of the STAV angle of attack and sideslip angle.

[0094] The feasible region is divided into several parts using straight lines parallel to the horizontal axis. To maintain continuity, oblique lines are used to connect the straight lines to the feasible region sequentially. The intersections form several trapezoidal domains, as shown in the attached diagram. Figure 2 As shown. The trapezoidal domain can be described by the following inequality constraints.

[0095] ①If ,at this time ;

[0096] ②If ,at this time ;

[0097] ③If ,at this time ;

[0098] ④If ,at this time .

[0099] in, and Let these be the coordinates of the right vertices of each trapezoidal domain. The trapezoidal domain description control law is simple and feasible to design, ensures the continuity of the feasible region, and provides a more accurate representation. Increasing the number of trapezoidal domains can improve the accuracy of the feasible region description. The accuracy of the description.

[0100] 2.3 Design of a dual-channel coupled command filter

[0101] Define the following dual-channel coupled command filter

[0102] (2)

[0103] Here, the subscript c represents the control signal, and the control signal c is filtered by the command filter to obtain the command signal d that the controller can follow. Represents the hyperbolic cosine function. Represents the hyperbolic tangent function. and The compensation parameters for the filter to be designed are as follows: and Convergence parameters of the filter to be designed. and The difference between them will eventually be compensated by the saturation signal. Feedback to the outer loop translation subsystem can then correct the track angle error. This is because the error constraint has a steady-state value. Within the safety constraints, the command constraints must leave room for error constraints; therefore, the command signal... With virtual control input Satisfy the following inequalities

[0104] (3)

[0105] in, These represent the maximum allowed values ​​for the angle of attack and sideslip angle. For specific non-rectangular domain safety constraints, the upper and lower bounds of the sideslip angle safety constraint are functions of the angle of attack, as shown in the attached figure. Figure 2 ,at this time, According to the trapezoidal description method, the safety domain is divided into zones within the angle of attack range. The intersections of the horizontal line and the safety domain boundary are denoted from bottom to top as follows: ,but

[0106] (4)

[0107] From the above, it can be concluded that when the safety domain is fixed, the smaller the steady-state value of the error constraint, the larger the allowable range of the command signal. During simulation verification, as long as the maximum range can be tracked using the preset error constraint control method... This ensures stable tracking of any command signal within the feasible region. Ultimately, the STAV's angle of attack and sideslip angle can be constrained through command signal constraints and tracking error constraints.

[0108] 2.4 Dual-channel coupling control constraints

[0109] When the error constraint steady-state value When the value is large, the dual-channel coupled command filter allows A smaller angle of attack and sideslip is detrimental to the performance of the STAV controller. In the attitude controller design process, in addition to limiting the state range of the angle of attack and sideslip angle, it is also necessary to constrain the transient and steady-state performance of the tracking error. Therefore, this invention also defines error constraints, performance constraints, and control constraints to facilitate the design of the coupled funnel controller.

[0110] (1) Error Constraint: Define the airflow angle tracking error as follows: , and Assume the airflow angle error satisfies the following error constraints.

[0111] (5)

[0112] Among them, the error envelope function for

[0113] (6)

[0114] In the formula, Here are the parameters to be set, representing the initial value of the performance envelope and the convergence rate of the performance envelope exponential, respectively. This represents the steady-state value constrained by the error.

[0115] (2) Performance constraints (or soft constraints): Performance constraints can be defined based on error constraints.

[0116] (7)

[0117] in, , Then there is , .

[0118] (3) Security Constraint (or Hard Constraint): Based on the analysis of the STAV dual-channel coupled constraint problem, the security constraint is defined as follows:

[0119] (8)

[0120] in, , ,and , The dual-channel coupled security constraints can be mathematically characterized using the trapezoidal domain description method.

[0121] For the STAV attitude control problem, the feasible region of the safety constraints is larger than the range that the error constraints can achieve, i.e.

[0122] (9)

[0123] (4) Control-oriented Constraint: Based on performance constraints and safety constraints, define control constraints.

[0124] (10)

[0125] in, The upper and lower bounds satisfy

[0126] (11)

[0127] And compensation factor satisfy

[0128] (12)

[0129] in, The parameters to be designed must satisfy the following inequalities.

[0130] (13)

[0131] Based on the control constraints, the updated control error constraints can be obtained.

[0132] (14)

[0133] If the system state can be made to satisfy the control constraint (10) by designing a control law, then the control constraint has the following three possible forms, taking the angle of attack control constraint as an example:

[0134] ①When hour, ,but

[0135] (15)

[0136] As attached Figure 3 As shown in (a), (b), (c), (f) and (g).

[0137] ②When hour, ,but

[0138] (16)

[0139] As attached Figure 3 As shown in (d) and (e).

[0140] ③When hour, ,but

[0141] (17)

[0142] As attached Figure 3 As shown in (h) and (i).

[0143] Step 3: Design the Coupled Funnel Controller

[0144] The flowchart for designing a STAV attitude controller based on a coupled command filter is attached. Figure 4 As shown.

[0145] To address the STAV attitude control problem with both safety and performance constraints, and to achieve stable and real-time attitude control, this invention decomposes the problem into two stages: attitude control and control allocation. The control constraint for the airflow roll angle is further defined as follows: If the design control law is based on dual-channel coupled control constraints, then the control constraints for the angle of attack and sideslip angle are defined as Equation (10); if the design control law is based on dual-channel coupled command filter, then the control constraints for the angle of attack and sideslip angle are defined as...

[0146] (18)

[0147] Regardless of which definition method is used for the control constraints, it will not affect the subsequent attitude controller design. This is to achieve the desired airflow angle command signal. For servo tracking, the STAV attitude angle differential equation is rewritten as the following tracking error differential equation.

[0148] (19)

[0149] in, For the virtual angular velocity command to be designed, To calculate the tracking error between control levels, The three-axis torques to be assigned, For torque distribution error, Let be the virtual torque coefficient to be assigned. Then, the control objective in this invention is equivalent to:

[0150] ① Based on the concept of nonlinear dynamic inverse, a dual-channel coupled funnel controller is designed. With the control distribution law, the STAV airflow angle is adjusted. Able to track stable command signals In a closed-loop system, all signals are semi-globally consistent and bounded, and the actuator... Satisfy the amplitude and bandwidth constraints;

[0151] ②The angle of attack and sideslip angle of the STAV satisfy the safety constraints of the non-rectangular domain;

[0152] ③ Under the condition of meeting safety constraints, the airflow angle tracking error of STAV and It can meet the pre-defined performance constraints.

[0153] 3.1 Attitude Control

[0154] Step 1: Define standardized state variables

[0155] (20) Equivalent to standardized error variable

[0156] (twenty one)

[0157] Limited time Define the conversion error within the range.

[0158] (twenty two)

[0159] Design NDI control law

[0160] (twenty three)

[0161] in, For virtual control input, It is a positive definite diagonal matrix. This represents the error transformation function. The proportional gain to be designed for each channel. This represents the standardized error variable. The first derivative over time can be obtained

[0162] (twenty four)

[0163] in and

[0164] (25)

[0165] Step 2: Define the angular rate tracking errors as follows: , and ,in, These are the roll, pitch, and yaw rates, respectively. Assume the airflow angle error satisfies the following error constraints.

[0166] (26)

[0167] Wherein, the error envelope function is

[0168] (27)

[0169] In the formula, These are parameters to be set. The steady-state value is defined as the error constraint. The control constraint for the rate of change of angular velocity is defined.

[0170] (28)

[0171] Define standardized error variables

[0172] (29)

[0173] Define conversion error

[0174] (30)

[0175] Design the control law (written in matrix form).

[0176] (31)

[0177] in, For virtual control input torque, It is a positive definite diagonal matrix. For The first derivative over time can be obtained

[0178] (32)

[0179] in and

[0180] (33)

[0181] As long as the transformation state is achieved through the design of the control law... In a limited time If the range is bounded, then the state The control constraints are satisfied.

[0182] 3.2 Torque Control Distribution

[0183] For the STAV rudder effect system, there are significant aerodynamic coupling characteristics between the actuator and the fuselage, and between actuators themselves. The relationship between the torque coefficient and the rudder deflection angle is not a simple linear one, limiting the applicability of control allocation methods based on affine models. To improve the applicability of control allocation methods, this invention is based on a non-affine rudder effect model. Design a control assigning law. Since non-affine functions do not explicitly contain control inputs, one approach to this problem is to linearize the non-affine function at the equilibrium point, transforming it into a function that linearizes at the current state. With control input Linearization is performed. The specific method is as follows: Linearize the non-affine function... exist Perform a first-order Taylor expansion at this point:

[0184] (34)

[0185] in, This represents the current actuator deflection. for and The higher-order infinitesimal. Control assignment algorithms used for flight control typically operate at a frequency of 100Hz, with sufficiently small time intervals. According to the time-scale separation principle, the state... Relative to control surface deflection As it is a slow variable, the state can be processed within one step of computation time. Treat it as a constant, that is Simultaneously, the torque coefficient increment at adjacent time points can be calculated. Simplified to the following form: Jacobian matrix multiplied by the manipulator increment:

[0186] (35)

[0187] Incremental rudder effect matrix for

[0188] (36)

[0189] Partial derivatives Aerodynamic parameters are obtained in real time using the central difference method.

[0190] (37)

[0191] in, In addition to control input Besides, the vector consists of the remaining input variables. To ensure sufficient assignment accuracy, let... During simulation, angles must be standardized to radians. In practical systems, actuators typically have amplitude and bandwidth limitations. The amplitude and speed limitations of aerodynamic rudders can be expressed in the following incremental form.

[0192] (38)

[0193] in, The sampling step size is typically a fixed 0.01. This completes the reconstruction of the control allocation problem in incremental form, with the solution range being the intersection of the rudder deflection amplitude and the bandwidth constraint. To ensure the uniqueness of the solution, the torque distribution error can be considered. Considering total rudder deflection only on the minimum basis. Minimum. The STAV control assignment problem can be solved using the following quadratic programming approach.

[0194] (39)

[0195] in, It is the control input when the control surface is at zero deflection. This is the current control input. It is the difference between the next time step and the current rudder deflection angle, which is the quantity to be designed; and The objective function consists of two parts, representing the torque coefficient generated by the current rudder deflection angle and the torque coefficient required to track the attitude command signal, respectively: The aim is to minimize torque distribution error; Aimed at minimizing overall rudder offset, Let be the weight matrix. Finally, equation (39) can be solved using mathematical programming. In summary, the incremental control allocation problem is to solve the rudder deflection increment using the effective set quadratic programming algorithm based on performance indicators and constraints. .

[0196] It should be noted that, in addition to the parameters explained above, the meanings of other parameters involved in this invention are shown in Table 2 below.

[0197] Table 2 Parameter Definitions

[0198]

[0199] Simulation experiment:

[0200] To verify the effectiveness of the dual-channel coupled funnel control method proposed in this invention, two simulation experiments were set up: "high-altitude high-speed penetration" and "low-altitude low-speed penetration." The simulation object was the STAV short-period three-channel wind tunnel data model, and all model parameters were consistent with those in the literature [Niestroy MA, Dorsett KM, Markstein K A. Taillessfighter aircraft model for control-related research and development[C]. AIAASciTech Forum, 2017, doi: 10.2514 / 6.2017-1757.]. It was assumed that the safe domains for the angle of attack and sideslip angle were... Figure 5 The non-rectangular domains are shown. The sideslip angle and angle-of-attack safety domains for the two sets of mission scenarios have similar shapes, but their sizes differ due to variations in flight missions and environments. When the angle of attack is too large or too small, the allowable sideslip angle range narrows: the larger the angle of attack, the lower the aerodynamic efficiency of the all-moving wingtip at the rear of the fuselage, making it difficult to correct for excessively large sideslip angles; the smaller the angle of attack, the lower the inlet efficiency, and the smaller the allowable sideslip angle. Trapezoidal descriptions are used to describe these safety constraints in the simulations.

[0201] Simulation 1: High-altitude penetration

[0202] To simulate a STAV high-altitude penetration mission, set the conditions. Using the traditional NDI method as the control group (Case 1), the proposed funnel control method based on a dual-channel coupled instruction filter (CFF) in this invention was used as the experimental group (Case 2), and simulations were performed for comparison. Since the feasible region is related to... Because it is symmetrical, only the key coordinates (unit: deg) of the right side of the trapezoidal domain are given.

[0203]

[0204] The remaining simulation parameters are set as follows:

[0205] (1) Parameter settings for NDI Case 1:

[0206] ① Control Law:

[0207] (40)

[0208] (41)

[0209] ② Adaptive parameters: ;

[0210] ③ Control law proportional gain: ;

[0211] (2) CFF Case 2 parameter settings:

[0212] ① Control laws: Equations (23) and (31);

[0213] ② Error envelope function:

[0214] (42)

[0215] Control law proportional gain: .

[0216] Appendix Figure 6 - Appendix Figure 11 These are the simulation results for STAV attitude control, Case 1 and Case 2. (Attached is...) Figure 6 For the STAV attitude angle tracking performance results, from Figure 6 From (a), we can see that even virtual control input Outside the security domain, command signals within the security domain can also be obtained through a coupled command filter. Furthermore, a margin was left for error constraints; Case 1, due to the lack of consideration for error constraints, resulted in actual sideslip angles and angles of attack exceeding the safety constraints, while Case 2's actual sideslip angles and angles of attack remained within the safety constraints. (See Appendix) Figure 6 As can be seen in (b), the airflow roll angle can be stabilized near 0 degrees. (Appendix) Figure 6 (c) and (d) show that in Case 1, the actual angle of attack and sideslip angle of STAV cannot meet the safety and performance constraints; in Case 2, the performance constraint range is always within the safety constraint range, and the angle of attack and sideslip angle can not only stably track the command signal but also meet the safety and performance constraints at the same time. When the angle of attack increases, the safety constraint of the sideslip angle automatically tightens, which reflects the coupling characteristics between channels.

[0217] Appendix Figure 7 The results of STAV attitude angle and angular rate tracking errors are shown in the attached figure. Figure 7 As can be seen, both the airflow angle tracking error and the attitude angle angular rate tracking error satisfy the error constraints.

[0218] Appendix Figure 8 For the STAV attitude angular rate and torque results, from the attached... Figure 8 As can be seen, Case 1 used a larger proportional gain to achieve stable tracking. Nearby, the torque of Case 1 produced a larger initial jump than the torque of Case 2; and At that time, due to Due to a sudden change, in order to maintain the preset performance, both the attitude angle change rate and torque change curves of Case 2 showed a certain degree of chattering.

[0219] Appendix Figure 9 The results for the STAV aerodynamic control surface deflection angle are from the attached... Figure 9 As can be seen, all actuators satisfy the amplitude and bandwidth constraints. In a local time period, the control input for Case 2 is larger than that for Case 1, because higher tracking performance requirements typically require greater control energy.

[0220] Appendix Figure 10 The attached diagram shows the adaptive parameter variation curves for the controller in Case 1. Figure 11 For the Case 2 controller conversion error curve, from the attached... Figure 10 and attached Figure 11 It can be seen that the adaptive parameters in Case 1 and the conversion error in Case 2 are both bounded. Overall, compared with Case 1, Case 2 has better tracking performance, but in the region of sudden change in command signal, a larger control surface deflection is required to obtain the torque to maintain tracking accuracy.

[0221] Simulation 2: Low-altitude penetration

[0222] To simulate a low-altitude penetration mission, set the conditions. Using the traditional barrier Lyapunov functions (BLF) as the control group (Case 3), and the proposed control constraint-based funnel control (CCF) method (Case 4) as the experimental group, simulations were conducted for comparison. In Case 3, error constraints were not considered; only safety constraints were taken into account. To improve system stability, a proportional element was added, while the remaining parameters and control law settings were consistent with Case 4. Since the feasible region is related to… Because it is symmetrical, only the key coordinates (unit: degrees) of the right side of the trapezoidal domain are given.

[0223]

[0224] (1) Case 3 BLF parameter settings:

[0225] ① Control Law:

[0226] (43)

[0227] ② Control constraint envelope function: ;

[0228] ③ Control law proportional gain:

[0229] .

[0230] (2) Case 4 CCF parameter settings:

[0231] ① Control laws: Equations (23) and (31);

[0232] ② Error envelope function:

[0233] (44)

[0234] ③ Control law proportional gain: ;

[0235] Regulatory factors: .

[0236] Appendix Figures 12-16 These are simulation results for STAV attitude control Case 3 and Case 4. (Attached) Figure 12 For the STAV attitude angle tracking performance results, from the attached Figure 12 From (a), it can be seen that in Case 3 and Case 4, the STAV sideslip angle and angle of attack both satisfy the safety constraints. When the command signal... Within the security domain, Case 4 It can achieve stable tracking with a small error, when the command signal is in Outside the security domain, in Case 4 It can get as close as possible to the command signal within a safe range. Compared to Case 3, Case 4 has a smaller steady-state tracking error.

[0237] From the appendix Figure 12 (b) shows that the airflow roll angle can be stabilized near 0 degrees. (See attached diagram.) Figure 12 Figures (c) and (d) show that as the angle of attack increases, the sideslip angle safety constraint automatically tightens, reflecting the coupling characteristics between channels. In Case 3, the actual angle of attack and sideslip angle of STAV only satisfy the safety constraint but not the performance constraint. In Case 4, when the command signal is within the safe region, both the angle of attack and sideslip angle satisfy both the safety and performance constraints; when the command signal is outside the safe region, only the safety constraint is satisfied. To minimize tracking error, both the sideslip angle and angle of attack curves closely follow the edge of the safety constraint. (Appendix) Figure 12 The green areas in (e) and (f) represent the control constraint range that couples safety constraints and performance constraints. According to the simulation results, in Case 4, the STAV sideslip angle and angle of attack both satisfy the control constraints.

[0238] Appendix Figure 13 The results of STAV attitude angle and angular rate tracking errors are shown in the attached figure. Figure 13 It can be seen that when the command signal is within the safe zone, both the airflow angle tracking error and the attitude angle / angular rate tracking error in Case 4 meet the error constraints; however, when the command signal is outside the safe zone, the error constraints cannot be met. For example: when When the signal is outside the safe zone, i.e. , Outside the error constraint range, but The safety constraints are still met. Furthermore, the attitude angular rate tracking errors in both Case 3 and Case 4 meet the performance constraints.

[0239] Appendix Figure 14 For the STAV attitude angular rate and torque results, from the attached... Figure 14 As can be seen, in order to maintain the preset performance, compared with Case 3, the attitude angle change rate and torque change curves of Case 4 both showed a certain degree of chattering.

[0240] Appendix Figure 15 The STAV aerodynamic control surface deflection angle curve is shown in the attached figure. Figure 15 As can be seen, all actuators satisfy the amplitude and bandwidth constraints. Overall, the control input of Case 4 is larger than that of Case 3. This is because higher tracking performance requirements usually require more control energy.

[0241] Appendix Figure 16 For the Case 4 controller conversion error curve, from the attached... Figure 16 It can be seen that the conversion errors in Case 4 are all bounded.

[0242] In summary, this invention designs a funnel controller based on a coupled command filter and control constraints, enabling dual-channel coupled safety and performance constraints. These two controller design concepts are then applied to simulations of two typical mission scenarios: high-altitude penetration and low-altitude infiltration. The simulation results are summarized in Table 3, demonstrating the effectiveness of the controller designed in this invention.

[0243] Table 3. Simulation settings and conclusions of the STAV dual-channel coupled funnel control example.

[0244]

[0245] Example 2:

[0246] Example 2 provides a dual-channel coupled control system for a supersonic tailless aircraft, specifically including an attitude control model establishment module, a control constraint definition module, and a control module;

[0247] The attitude control model establishment module is used to establish the attitude control model of the supersonic tailless vehicle; the control constraint definition module is used to determine the control constraint conditions for the coupled control of the supersonic tailless vehicle; and the controller module is used to control the attitude and torque distribution of the supersonic tailless vehicle.

[0248] It should be noted that the attitude control model establishment module, control constraint definition module, and control module are all implemented based on the control method described in Embodiment 1.

[0249] Example 3:

[0250] Embodiment 3 provides a dual-channel coupled control device for a supersonic tailless aircraft, including at least one processor; and a memory communicatively connected to the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor to enable the processor to perform the control method described in Embodiment 1.

[0251] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A dual-channel coupled control method for a supersonic tailless vehicle, characterized in that, Includes the following steps, S1: Establish an attitude control model for a supersonic tailless aircraft; S2: Based on dual-channel coupling constraints, determine the control constraint conditions for the coupled control of a supersonic tailless vehicle; S3: Design a coupled funnel controller to control the attitude and torque distribution of a supersonic tailless aircraft; The specific operation of step S2 includes the following steps: S201: Determine the angle of attack of a supersonic tailless vehicle With sideslip angle Dual-channel non-rectangular domain security constraints are proposed, and a mathematical description of the non-rectangular domain security constraints is provided. S202: Design a dual-channel coupled command filter; S203: Based on performance constraints and safety constraints, define dual-channel coupled control constraints and determine control constraint conditions; The dual-channel coupled command filter in step S202 is ； Wherein, subscript c represents control signal and subscript d represents command signal; Represents the hyperbolic cosine function. Represents the hyperbolic tangent function. and The compensation parameters for the filter to be designed are as follows: and Convergence parameters of the filter to be designed.

2. The dual-channel coupled control method for a supersonic tailless vehicle according to claim 1, characterized in that: The attitude control model of the supersonic tailless aircraft mentioned in step S1 is as follows: ； In the formula, For the STAV system status, For aerodynamic torque, This is the aerodynamic torque coefficient; This is the transformation matrix from the body coordinate system to the airflow coordinate system. This is the transformation matrix from torque to attitude angle and angular rate. This is the transformation matrix from torque coefficient to torque; Indicates dynamic pressure. Indicates the reference area of ​​the aircraft; The vector composed of aerodynamic control surfaces; This is due to unknown external interference. It is a nonlinear function related to the system state; It is a non-affine function.

3. The dual-channel coupled control method for a supersonic tailless vehicle according to claim 2, characterized in that, The specific operation of step S203 includes the following steps: S2031: Define the airflow angle tracking error as follows: , and Then the airflow angle error satisfies the error constraint. ； In the formula, Let be the error envelope function; S2032: Performance constraints are defined based on error constraints. ； in, , ; S2033: Based on dual-channel coupling constraints, the security constraints are defined as follows: in, , ; S2034: Based on performance constraints and security constraints, the control constraints are defined as follows: ； in, , This is a compensation factor.

4. The dual-channel coupled control method for a supersonic tailless vehicle according to claim 3, characterized in that, Step S3 includes the following steps: S301: Rewrite the attitude control model of a supersonic tailless aircraft as a tracking error differential equation. ； in, This is an airflow angle command signal. For the virtual angular velocity command to be designed, To calculate the tracking error between control levels, The three-axis torques to be assigned, For torque distribution error, The virtual torque coefficient to be assigned; S302: Design of a Dual-Channel Coupled Funnel Controller With the torque control distribution law, the STAV airflow angle is adjusted. Able to track stable command signals .

5. The dual-channel coupled control method for a supersonic tailless vehicle according to claim 4, characterized in that, The steps described in step S302 The control law is ； The control law is ； In the formula, For virtual control input, Input torque for virtual control; These are the roll, pitch, and yaw rates, respectively. and It is a positive definite diagonal matrix. This represents the error transformation function. The proportional gain to be designed for each channel. This represents the standardized error variable.

6. The dual-channel coupled control method for a supersonic tailless vehicle according to claim 5, characterized in that, The torque control distribution law mentioned in step S302 is as follows: ； In the formula, It is the control input when the control surface is at zero deflection. This is the current control input. It is a non-affine function. It is the difference between the next time step and the current rudder deflection angle, which is the quantity to be designed; and The objective function consists of two parts, representing the torque coefficient generated by the current rudder deflection angle and the torque coefficient required to track the attitude command signal, respectively: The aim is to minimize torque distribution error; Aimed at minimizing overall rudder offset, The weight matrix, This is the incremental rudder effect matrix.

7. A dual-channel coupled control system for a supersonic tailless aircraft, characterized in that: It includes an attitude control model establishment module, a control constraint definition module, and a control module; The attitude control model establishment module is used to establish the attitude control model of the supersonic tailless aircraft. The control constraint definition module is used to determine the control constraint conditions for the coupled control of a supersonic tailless vehicle. The controller module is used to control the attitude and torque distribution of the supersonic tailless aircraft; The attitude control model establishment module, control constraint definition module, and control module are implemented based on the control method described in any one of claims 1-6.

8. A dual-channel coupled control device for a supersonic tailless aircraft, characterized in that, The device includes at least one processor; and a memory communicatively connected to the processor; wherein the memory stores instructions executable by the processor to enable the processor to perform the control method according to any one of claims 1-6.

Images