An automatic carrier landing control method and system

Abstract

The application discloses an automatic landing control method and system, relates to the field of flight control technology, and comprises the following steps: constructing a longitudinal dynamics model of an airplane, and discretizing the model to obtain a discrete state space equation; acquiring the longitudinal motion state quantity of the airplane, the steady-state target quantity of the deck of an aircraft carrier and the steady-state disturbance quantity of the wake flow of the aircraft carrier at the current moment; solving the discrete state space equation based on the equation by using a whale optimization algorithm to obtain an optimal prediction control input sequence; filtering and correcting the longitudinal motion state of the airplane after control by using a Kalman filter; and landing if the corrected state meets preset landing conditions, or returning to acquire data again. The application combines model prediction control dynamic constraints and Kalman filter noise reduction, and quickly solves by using a whale optimization algorithm, so that the landing precision and anti-interference ability in a complex environment are significantly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of flight control technology, in particular to an automatic landing control method and system. BACKGROUND

[0002] As the core force of sea operations, the automatic landing system of a carrier-based aircraft needs to achieve precise landing without human intervention in the whole process under complex marine environments (such as strong wind disturbance, deck movement, and sea wave impact) and sensor noise interference, which puts high requirements on the real-time performance, robustness, and anti-interference ability of the control system.

[0003] With the development of avionics technology and computer control technology, automatic landing control has become a key research field for ensuring the operational efficiency of carrier-based aircrafts. However, the automatic landing control method in the related art has significant technical defects: on the one hand, the state constraints and input constraints of the controlled object are not considered as optimization prerequisites, which leads to control input exceeding physical limits and causing control saturation and even system instability; on the other hand, although dynamic constraints can be handled, the influence of system uncertainties on the controlled object is not effectively integrated, which results in insufficient tracking accuracy of the ideal landing point, especially a significant decrease in the success rate of hooking under complex sea conditions.

[0004] Therefore, there is an urgent need for an automatic landing control method to improve the landing accuracy and system robustness of carrier-based aircrafts in complex environments. SUMMARY

[0005] The present application aims to provide an automatic landing control method and system that can effectively solve the problems of insufficient handling of dynamic constraints and weak suppression of uncertainties during automatic landing of carrier-based aircrafts in complex marine environments (such as strong wind disturbance, carrier wake disturbance, and sensor noise interference), achieve high-precision tracking of the ideal landing point, and improve the landing success rate and system robustness of carrier-based aircrafts in dynamic constraint and strong interference coupled scenarios.

[0006] To achieve the above-mentioned purpose, the present application provides the following solutions:

[0007] In a first aspect, the present application provides an automatic landing control method, comprising:

[0008] Constructing a longitudinal dynamics model of the aircraft and discretizing it to obtain a discrete state space equation for describing the longitudinal motion of the aircraft;

[0009] Obtaining the longitudinal motion state quantity of the aircraft, the steady-state target quantity of the aircraft carrier deck, and the steady-state disturbance quantity of the carrier wake at the current time;

[0010] The longitudinal motion state quantity of the aircraft at the current time is taken as an initial motion state, and the optimal predictive control input sequence is obtained by solving the discrete state space equation based on a whale optimization algorithm; the control input includes a rudder deflection angle of an elevator, a rudder deflection angle of a flap and a deflection angle of a throttle lever;

[0011] The longitudinal motion state of the aircraft obtained by controlling the aircraft based on the first control input in the optimal predictive control input sequence is filtered by a Kalman filter to obtain a corrected longitudinal motion state of the aircraft;

[0012] If the corrected longitudinal motion state of the aircraft meets a preset landing condition, automatic landing is performed; the preset landing condition includes that a vertical height error between the aircraft and a target landing point is less than a preset height threshold and a speed error is less than a preset speed threshold;

[0013] If the corrected longitudinal motion state of the aircraft does not meet the preset landing condition, the step of obtaining the longitudinal motion state quantity of the aircraft at the current time, the carrier deck motion steady target quantity and the disturbance steady quantity of the carrier stern flow is returned to.

[0014] In a second aspect, the present application provides an automatic landing control system, comprising:

[0015] A model discretization module is configured to construct an aircraft longitudinal dynamics model and perform discretization to obtain a discrete state space equation for describing the longitudinal motion of the aircraft;

[0016] A state quantity acquisition module is configured to obtain the longitudinal motion state quantity of the aircraft at the current time, the carrier deck motion steady target quantity and the disturbance steady quantity of the carrier stern flow;

[0017] A control sequence solving module is configured to take the longitudinal motion state quantity of the aircraft at the current time as an initial motion state, and obtain the optimal predictive control input sequence by solving the discrete state space equation based on a whale optimization algorithm; the control input includes a rudder deflection angle of an elevator, a rudder deflection angle of a flap and a deflection angle of a throttle lever;

[0018] A Kalman filtering module is configured to filter the longitudinal motion state of the aircraft obtained by controlling the aircraft based on the first control input in the optimal predictive control input sequence by a Kalman filter to obtain a corrected longitudinal motion state of the aircraft;

[0019] A landing judgment module is configured to perform automatic landing if the corrected longitudinal motion state of the aircraft meets a preset landing condition; the preset landing condition includes that a vertical height error between the aircraft and a target landing point is less than a preset height threshold and a speed error is less than a preset speed threshold;

[0020] The feedback control module is configured to return to the step of obtaining the longitudinal motion state quantity of the aircraft, the steady-state target quantity of the deck of the aircraft carrier and the steady-state disturbance quantity of the wake of the aircraft carrier at the current time if the modified longitudinal motion state of the aircraft does not satisfy the preset carrier landing condition.

[0021] In a third aspect, the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the automatic carrier landing control method according to any one of the preceding aspects.

[0022] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the automatic carrier landing control method according to any one of the preceding aspects.

[0023] In a fifth aspect, the present application provides a computer program product, comprising a computer program, and the computer program is executed by a processor to implement the steps of the automatic carrier landing control method according to any one of the preceding aspects.

[0024] According to the embodiments provided in the present application, the present application has the following technical effects:

[0025] The present application provides an automatic carrier landing control method and system, constructs an aircraft longitudinal dynamics model and performs discretization processing, obtains a discrete state space equation capable of accurately describing the longitudinal motion characteristics of the aircraft, solves the problem of insufficient modeling accuracy caused by low real-time calculation efficiency and dynamic response lag of the traditional continuous model in a complex marine environment, realizes high-precision and high-real-time dynamic modeling of the longitudinal motion state of the carrier-based aircraft, and provides a reliable basic model support for subsequent control strategies; the current time aircraft longitudinal motion state quantity is taken as the initial state, and the whale optimization algorithm is used to solve the prediction control problem based on the discrete state space equation, to obtain an optimal prediction control input sequence containing the elevator deflection angle, the flap deflection angle and the throttle lever deflection angle, solve the problem of control input saturation and insufficient trajectory tracking accuracy of the traditional control method in the coupling scene of dynamic constraints (such as physical limitations of the control surface and the thrust range of the engine) and strong uncertain disturbances (such as sea wave impact and wake disturbance), and realize real-time trajectory optimization and robust control under multiple constraint conditions; the Kalman filter is used to filter the aircraft longitudinal motion state obtained after the first control input in the optimal prediction control input sequence is implemented, effectively suppresses the influence of system uncertainty and measurement noise on state estimation, solves the problem of control performance degradation caused by state information distortion in the traditional feedback control, realizes the stability improvement of the high-precision state feedback and closed-loop control system, and finally significantly improves the automatic carrier landing success rate, trajectory tracking accuracy and system robustness of the carrier-based aircraft in a complex marine environment. BRIEF DESCRIPTION OF DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed in the embodiments will be briefly introduced as follows. Obviously, the accompanying drawings in the following description only relate to some embodiments of the present application, and for those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0027] Figure 1 A flowchart of an automatic landing control method according to an embodiment of the present application.

[0028] Figure 2 A flowchart of an automatic landing control method according to another embodiment of the present application.

[0029] Figure 3 A landing height error comparison curve diagram according to an embodiment of the present application.

[0030] Figure 4 A reference speed error comparison curve diagram in a landing process according to an embodiment of the present application.

[0031] Figure 5 A reference angle of attack error comparison curve diagram in a landing process according to an embodiment of the present application.

[0032] Figure 6 An elevator deflection angle change curve diagram in a landing process according to an embodiment of the present application.

[0033] Figure 7 A reference throttle deflection angle change curve diagram in a landing process according to an embodiment of the present application.

[0034] Figure 8 A functional module diagram of an automatic landing control system according to an embodiment of the present application. DETAILED DESCRIPTION

[0035] First, some technical terms involved in the embodiments of the present application are introduced.

[0036] As a core component of the sea operation force, the automatic landing system of the carrier-based aircraft has become a research and application focus field along with the rapid development of communication technology and computer technology in recent years. The system aims to enable the carrier-based aircraft to automatically complete the landing task in the whole process without too much human intervention in the complex environment. This puts forward very high requirements on the performance of the controller, which needs to realize accurate tracking of the ideal landing point of the deck in the complex environment such as wind disturbance and observation sensor error disturbance under the restriction conditions of the output and state of the carrier-based aircraft, so as to ensure the safe and accurate landing of the carrier-based aircraft.

[0037] There are relevant researchers who propose an automatic landing control system based on the height of the active disturbance rejection control (ADRC), but the ADRC technology is suitable for SISO (Single-Input Single-Output) system, and the control design of the inner loop and the outer loop of the aircraft attitude needs to adjust more parameters. There are relevant researchers who propose a landing control scheme combining neural network and disturbance observer, use neural network system to process system uncertainty, approach external disturbance suffered by the aircraft through disturbance observer, and design landing control system based on output feedback. There are relevant researchers who propose a landing control scheme based on knowledge-guided optimization algorithm, improve the tracking accuracy of the aircraft through iterative optimization of the parameters of the dynamic inverse controller, and the system has good robustness. But the state of the controlled object in the controller of the above related technologies is not limited and constrained as a control condition prerequisite, and the problem of control saturation is easy to appear in the control process. The model predictive control (MPC) method is to predict the system state and input based on state limit and constraint condition, and the optimal solution is obtained by equation solving, and some articles have proposed to use MPC method to solve the automatic landing problem.

[0038] There are relevant researchers who propose a problem of using predictive control to suppress the wake disturbance during landing, and verify that the robustness of the landing control system under predictive control is good. There are relevant researchers who propose an MPC-LQG (Model Predictive Control-Linear Quadratic Gaussian) algorithm, which combines model predictive control and linear quadratic Gaussian algorithm controller to realize the accurate landing scheme of the aircraft in the longitudinal direction. But the above mentioned related technical solutions do not consider the influence of the uncertainty existing in the system on the controlled object, including the disturbance in the landing process and the noise of the observation sensor and other hardware devices, and the Kalman filter as a digital fusion algorithm has the characteristics of filter and observer, and plays a good role in noise reduction.

[0039] In order to ensure the requirement of accurate landing of the carrier-based aircraft in the whole process, a Kalman filter and a model predictive controller are fused (KF-MPC, Kalman Filter-Model Predictive Control), the state estimation provided by the Kalman filter is taken as the feedback value of the system controller, the output sequence is obtained through the rolling optimization, and then the optimal control is obtained through calculation iteration. However, since the output is provided by the Kalman filter, the commonly used convex point method for calculating the quadratic programming problem may have the problems of being unable to find the optimal solution and the long calculation convergence time. The whale optimization algorithm (WOA) can achieve fast convergence and global optimization of function optimization, so the controller in the application adopts the WOA optimization algorithm for solving.

[0040] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0041] In order to make the above-mentioned purposes, characteristics and advantages of the application more obvious and easy to understand, the application will be further described in detail below with reference to the drawings and specific embodiments.

[0042] In one exemplary embodiment, as shown in Figure 1 An automatic landing control method is provided, including the following steps 201 to 206. Among them:

[0043] Step 201, a longitudinal dynamics model of the aircraft is constructed, and is discretized to obtain a discrete state space equation for describing the longitudinal motion of the aircraft.

[0044] Step 202, the longitudinal motion state quantity of the aircraft at the current time, the steady-state target quantity of the aircraft carrier deck motion and the steady-state disturbance quantity of the aircraft tail flow are obtained.

[0045] Step 203, the longitudinal motion state quantity of the aircraft at the current time is taken as the initial motion state, and based on the discrete state space equation, the whale optimization algorithm is used for solving to obtain an optimal predicted control input sequence; the control input includes the rudder angle of the elevator, the rudder angle of the flap and the deflection angle of the throttle lever.

[0046] Step 204, the longitudinal motion state of the aircraft obtained by controlling the aircraft based on the first control input in the optimal predicted control input sequence is filtered by the Kalman filter to obtain the corrected longitudinal motion state of the aircraft.

[0047] If the modified longitudinal motion state of the aircraft meets the preset landing condition, the automatic landing is performed in step 205; the preset landing condition includes that a vertical height error between the aircraft and the target landing point is less than a preset height threshold and a speed error is less than a preset speed threshold.

[0048] If the modified longitudinal motion state of the aircraft does not meet the preset landing condition, the step of "obtaining the longitudinal motion state quantity of the aircraft, the motion steady-state target quantity of the aircraft carrier deck and the steady-state disturbance quantity of the ship wake flow at the current time" is returned to in step 206.

[0049] By implementing the above steps 201 to 206, the application can effectively solve the problems of insufficient dynamic constraint processing and weak uncertainty interference suppression faced by the carrier aircraft during automatic landing in a complex marine environment. Through accurate model construction, real-time data acquisition, optimized control input solution, and accurate state estimation and judgment, high-precision trajectory tracking and safe automatic landing of the carrier aircraft in the dynamic constraint and strong interference coupling scene are realized, which significantly improves the success rate of the carrier aircraft landing and the robustness of the system, and provides reliable technical support for the safe landing of the carrier aircraft.

[0050] In another exemplary embodiment of the application, step 201 specifically includes:

[0051] The longitudinal dynamics model of the aircraft is constructed by the following formula:

[0052] .

[0053] .

[0054] .

[0055] wherein, represents the flight acceleration; represents the mass of the aircraft; represents the engine thrust; represents the angle of attack; represents the flight resistance; represents the gravity acceleration; represents the glide angle; represents the pitch angle; represents the change rate of the glide angle; represents the flight speed; represents the lift; represents the pitch angle acceleration; represents the body pitch moment of inertia; represents the pitch moment; represents the change rate of the vertical height of the aircraft; represents the dynamic pressure; represents the wing area; Indicates the lift coefficient; Indicates zero lift coefficient; This represents the coefficient representing the effect of the angle of attack on the lift coefficient. The coefficient representing the effect of the elevator deflection angle on the lift coefficient; Indicates the elevator deflection angle; The coefficient representing the influence of the flap deflection angle on the lift coefficient; Indicates the rudder deflection angle of the flaps; Indicates the drag coefficient; Indicates zero drag coefficient; The coefficient representing the squared term of the lift coefficient; This represents the average chord length of the wing; Indicates the pitch moment coefficient; Indicates the pitching moment coefficient at zero angle of attack; This represents the coefficient representing the influence of the angle of attack on the pitching moment coefficient; Indicates pitch angular velocity; This represents the coefficient that indicates the influence of the elevator deflection angle on the pitch moment coefficient. The coefficient representing the influence of the flap deflection angle on the pitching moment coefficient; This represents the Mach number coefficient for throttle. Indicates the Mach number at throttle position; Indicates the engine height coefficient; Indicates the vertical altitude of the aircraft; This indicates the deflection angle of the throttle lever.

[0056] Based on the longitudinal dynamics model of the aircraft, the state vector is selected. With control input The equations are then discretized to obtain the discrete state-space equations describing the longitudinal motion of the aircraft:

[0057] .

[0058] in, This represents the predicted state vector at time k+1; Represents the state transition matrix; Represents the control input matrix; This represents the state vector at time k; This represents the control input at time k.

[0059] In another exemplary embodiment of this application, the steady-state target quantities of the carrier deck motion in step 202 include the heave displacement of the carrier's center of gravity and the pitch angle of the carrier deck.

[0060] The steady-state target quantity of the aircraft carrier deck motion is obtained using the following formula:

[0061] .

[0062] .

[0063] wherein, denotes the pitch angle of the aircraft carrier; denotes the current time; denotes the random disturbance noise of the pitch angle; denotes the heave amplitude of the center of gravity of the aircraft carrier; denotes the random disturbance noise of the heave displacement; denotes the motion amplitude of the target landing point; denotes the horizontal position of the target landing point on the deck; denotes the horizontal position of the center of gravity of the aircraft carrier.

[0064] The stern flow disturbance steady-state quantity comprises a stern flow horizontal disturbance steady-state quantity and a stern flow vertical disturbance steady-state quantity.

[0065] The stern flow disturbance steady-state quantity is obtained by the following formula:

[0066] .

[0067] wherein, denotes the stern flow horizontal disturbance steady-state quantity; denotes the steady-state wind component of the stern flow horizontal disturbance steady-state quantity; denotes the periodic disturbance component of the stern flow horizontal disturbance steady-state quantity; denotes the sea surface atmospheric turbulence disturbance component; denotes the stern flow vertical disturbance steady-state quantity; denotes the steady-state wind component of the stern flow vertical disturbance steady-state quantity; denotes the periodic disturbance component of the stern flow vertical disturbance steady-state quantity; denotes the random disturbance component of the stern flow vertical disturbance steady-state quantity.

[0068] In another exemplary embodiment of the present application, step 203 specifically comprises:

[0069] According to the discrete state space equation, a performance index function of a quadratic programming problem of the deviation between the predicted state sequence and the reference state sequence is constructed.

[0070] The difference between the sum of the longitudinal motion state quantity of the aircraft and the steady-state disturbance term and the expected state is taken as an error value, and the longitudinal motion state quantity of the aircraft in the performance index function is replaced to obtain a quadratic programming problem containing steady-state disturbance; the steady-state disturbance term comprises a steady-state disturbance of the steady-state target quantity of the aircraft carrier deck motion and a steady-state disturbance of the stern flow disturbance steady-state quantity.

[0071] The whale optimization algorithm is used to solve a quadratic programming problem containing steady-state disturbance to obtain an optimal predictive control input sequence.

[0072] In another example embodiment of the present application, the whale optimization algorithm is used to solve a quadratic programming problem containing steady-state disturbance to obtain an optimal predictive control input sequence, specifically comprising:

[0073] The positions of a whale population are initialized, and the initial value of the iteration number t is set to 1; wherein the position of each whale individual in the whale population corresponds to a candidate control input.

[0074] The fitness value of each whale individual in the whale population in the tth iteration is calculated; the fitness value is calculated based on the performance index function.

[0075] The position of the current optimal whale individual is determined according to the fitness value.

[0076] Based on the position of the current optimal whale individual, the positions of other whale individuals in the whale population except the current optimal whale individual are updated according to a preset update strategy; the preset update strategy includes a shrink-and-spiral mechanism update strategy and a spiral attack mode update strategy.

[0077] t is set to t+1, and the step of calculating the fitness value of each whale individual in the whale population in the tth iteration is returned until the maximum iteration number is reached, and the control input sequence corresponding to the position of the current optimal whale individual is output as the optimal predictive control input sequence.

[0078] In another example embodiment of the present application, based on the position of the current optimal whale individual, the positions of other whale individuals in the whale population except the current optimal whale individual are updated according to a preset update strategy, specifically comprising:

[0079] The position update control coefficient vector of the current iteration number is determined.

[0080] The scheme for calculating the position of the whale individual is determined according to the position update control coefficient vector and the generated random number, and the current whale individual position is updated.

[0081] If the random number is less than the preset probability threshold value, and the absolute value of the position update control coefficient vector is less than the preset absolute value threshold value, the shrink-and-spiral mechanism update strategy is used to update the positions of other whale individuals in the whale population except the current optimal whale individual.

[0082] If the random number is less than the preset probability threshold value, and the absolute value of the position update control coefficient vector is greater than or equal to the preset absolute value threshold value, the random selection search agent strategy is used to update the positions of other whale individuals in the whale population except the current optimal whale individual.

[0083] If the random number is greater than the preset probability threshold, a spiral attack mode update strategy is used to update the positions of the other whale individuals in the whale population except the current optimal whale individual.

[0084] In another exemplary embodiment of the present application, for the research on automatic carrier deck landing under interference, an automatic landing control method is proposed by fusing model predictive control and Kalman filter. First, the aircraft dynamics model is established and the interference of the landing environment is introduced, and the model predictive control is adopted to adapt to the dynamic planning and strict constraint of the controlled object. Second, based on the random interference of the environment and sensors, Kalman filter is introduced to remove noise, and the state estimation value of Kalman filter is taken as the feedback input of the controller to reduce the influence of uncertain noise. For fast calculation, the whale optimization algorithm (WOA) is used for controller calculation. Finally, simulation is carried out, and data comparison shows that the controller method based on the fusion of Kalman filter and model predictive control whale optimization realizes the effect of anti-interference and precise landing. As Figure 2 shown, an automatic landing control method is provided, which specifically comprises:

[0085] Step 1, landing model establishment.

[0086] 1.1, aircraft dynamics model establishment. During the process of aircraft glide landing, there are relatively fixed attitude requirements, but due to the interference of ship tail wind and deck movement, the state of the aircraft produces errors, and the external disturbance, especially the wind disturbance, has a great influence on the longitudinal motion of the aircraft. Therefore, the present application establishes a longitudinal dynamics model of the aircraft, as shown in formula (1):

[0087] (1).

[0088] wherein, represents flight acceleration; represents the mass of the aircraft; represents engine thrust; represents the angle of attack; represents flight resistance; represents gravity acceleration; represents glide angle; represents pitch angle; represents glide angle change rate; represents flight speed; represents lift; represents pitch angle acceleration; represents the moment of inertia of the body in pitch; represents the pitch moment; represents the rate of change of the vertical height between the aircraft and the target landing point.

[0089] , , The expressions of the above are respectively: (2).

[0090] wherein in formula (2) represents the dynamic pressure, represents the wing area, represents the mean chord length of the wing, are dimensionless lift, drag and pitching moment coefficients, and the expressions are respectively:

[0091] (3).

[0092] wherein and respectively represent the rudder deflection angle of the elevators and the flap deflection angle of the flaps; is the pitch angular velocity; represents the lift coefficient; represents the zero lift coefficient; represents the influence coefficient of the angle of attack on the lift coefficient; represents the influence coefficient of the rudder deflection angle of the elevators on the lift coefficient; represents the rudder deflection angle of the elevators; represents the influence coefficient of the flap deflection angle of the flaps on the lift coefficient; represents the flap deflection angle of the flaps; represents the drag coefficient; represents the zero drag coefficient; represents the square term coefficient of the lift coefficient; represents the mean chord length of the wing; represents the pitching moment coefficient; represents the zero angle of attack pitching moment coefficient; represents the influence coefficient of the angle of attack on the pitching moment coefficient; represents the influence coefficient of the rudder deflection angle of the elevators on the pitching moment coefficient; represents the influence coefficient of the flap deflection angle of the flaps on the pitching moment coefficient; engine thrust controlled by the throttle lever, and expressed as: (4). wherein represents the throttle Mach number coefficient; represents the throttle Mach number; represents the engine altitude coefficient; represents the vertical height between the aircraft and the target landing point; represents the throttle lever deflection angle.

[0093] ​​​​1.2, disturbance model. The aircraft carrier is affected by the wave and sea wind, resulting in 6 degrees of freedom of translation and rotation movement, in which the longitudinal pitch and heave motion is the most frequent, and the position of the landing point is the most affected. The amplitude and frequency of the motion under the influence of different sea wave motion are also quite different, especially in the case of high sea, the frequency and amplitude of the deck longitudinal motion are large, thereby reducing the success rate of the hook.

[0094] The simulation strategy of the aircraft carrier motion adopts the sine wave function method, which represents the pitch and heave motion of the deck by combining sine functions.

[0095] (5).

[0096] (6).

[0097] wherein, represents the pitch angle of the aircraft carrier; represents the current time; represents the random disturbance noise of the pitch angle; represents the heave motion amplitude of the center of gravity of the aircraft carrier; represents the random disturbance noise of the heave displacement; represents the motion amplitude of the target landing point; represents the horizontal position of the target landing point on the deck, and the target landing point is located at the middle point of the second and third arresting cables; represents the horizontal position of the center of gravity of the aircraft carrier. and represent the noise in the motion process.

[0098] The aircraft carrier wake model is established according to the American navy MIL-F-87857C military standard, the wake can be divided into longitudinal and vertical, and the horizontal wind is ignored, the expression is:

[0099] (7).

[0100] wherein, wherein, represents the horizontal disturbance steady-state quantity of the wake; represents the steady-state wind component of the horizontal disturbance steady-state quantity of the wake; represents the periodic disturbance component of the horizontal disturbance steady-state quantity of the wake; represents the sea surface atmospheric turbulence disturbance component; represents the vertical disturbance steady-state quantity of the wake; represents the steady-state wind component of the vertical disturbance steady-state quantity of the wake; represents the periodic disturbance component of the vertical disturbance steady-state quantity of the wake; represents the random disturbance component of the vertical disturbance steady-state quantity of the wake.

[0101] Step one finally gets Figure 2 the discrete system of the aircraft dynamic model, the deck motion steady-state target and the steady-state wind disturbance.

[0102] Step two, KF and MPC fusion optimization control design. The aircraft carrier landing controller design based on Kalman filter and model predictive control optimization fusion is shown in Figure 2 . First, the dynamic discrete system of the aircraft and the steady-state disturbance form an augmented matrix. Second, the model predictive controller is transformed into a quadratic programming form, and the WOA optimization algorithm is used to solve the quadratic programming problem. Then, due to the existence of noise, the predicted state is obtained after the Kalman filter, and the performance of the control system is improved by noise reduction to realize the precise landing of the aircraft.

[0103] The specific process is as follows:

[0104] Step 2.1, model predictive controller establishes system quadratic programming equation.

[0105] Based on the longitudinal dynamics model of the aircraft, the state vector and the control input are selected, and the discrete linear system state space equation at time k is expressed as:

[0106] (8).

[0107] wherein, represents the predicted state vector at the next time k+1; represents the state transition matrix; represents the control input matrix; represents the state vector at the current time k; represents the control input at the current time k.

[0108] Let the prediction interval length be , the state transition matrix is dimension, the control input matrix is dimension, and the predicted state sequence can be represented as:

[0109] (9).

[0110] wherein, represents the state transition cumulative matrix, which is composed of different powers of the state transition matrix A, and is used to describe the transition relationship of the state within the prediction interval; ; represents the control input influence distribution matrix; , which reflects the influence of control inputs on predicted states, combines the control input matrix B with different powers of the state transition matrix A, and is used to calculate the evolution of states under given control inputs.

[0111] , with dimension np x 1. N represents the dimension of system states, i.e., the total number of elements in the predicted state sequence .

[0112] represents a control input sequence; , with dimension mp x 1. m represents the dimension of control inputs, i.e., the total number of elements in the control input sequence .

[0113] The performance index of the quadratic form at the sampling time is expressed as:

[0114] (10).

[0115] wherein represents a performance index function; represents the total length of the prediction period; represents the predicted state vector at time k+p; T represents transposition; represents an input cost weight matrix; represents an end cost weight matrix; represents a prediction time; represents the predicted state vector at time ; represents the predicted control input at time ; represents a state cost weight matrix;

[0116] The initial state cost term in equation (10) is proposed, and the remaining cost terms can be written in the form of matrix multiplication, which can be expressed as:

[0117] (11).

[0118] wherein represents an extended state cost weight matrix, which integrates the end cost weight matrix and the state cost weight matrix S together to form a block diagonal matrix. Its form is , which is used to consider the cost of states at different times within the prediction interval uniformly in the cost function. represents an extended input cost weight matrix, which is a block diagonal matrix, and its form is The input cost weight matrix R is extended to match the dimension of the control input sequence to calculate the cost of the whole control input sequence in the cost function.

[0119] Next, the formula (9) is substituted into the formula (11) to obtain: Substitute The formula (9) is substituted into the formula (11) to obtain:

[0120] (12).

[0121] The first term is determined by the initial state and is irrelevant to the input, so the first term is ignored when the cost is optimized, and the standard form of the quadratic programming problem is obtained:

[0122] (13).

[0123] wherein, represents the initial state associated linear term coefficient vector, which is linearly related to the control input sequence by the interaction of the initial state and the system matrix; represents the comprehensive control cost quadratic term coefficient matrix, which is a symmetric matrix, and comprehensively reflects the influence of the control input on the state and the input cost weight, etc., determines the characteristics of the quadratic term in the cost function, and has an important influence on the solution of the quadratic programming problem and the properties of the optimal solution.

[0124] Step 2.2, quadratic equation introduces steady-state disturbance. In the simulation of the landing process, there is a steady-state motion problem in the tracking of the target landing point. The state of the aircraft in the performance index of formula (10) is replaced by the error value which is expressed as:

[0125] (14).

[0126] wherein, represents the error state vector at time k to time ; p represents the prediction interval length; represents the predicted state vector at time k to time ; represents the target steady-state motion state, i.e. the steady-state motion state that the aircraft is expected to reach, which is the target state to be tracked; represents the control input vector at time k to time ; is the augmented state coefficient matrix; represents the predicted state augmented matrix at time k to time ; the control term is added in the performance index, which is to make the system control more smooth and avoid the situation of excessive shaking of the system output. Therefore, the output in formula (10) ​​replaced with an output change amount The performance index is finally expressed as:

[0127] (15).

[0128] wherein, denotes the control output change amount at time k to time ; denotes the transformed cost weight matrix related to the state augmentation matrix; , and denote the extended state cost weight matrix and the extended input cost weight matrix after the augmentation processing, respectively; denotes the augmented state sequence; denotes the state augmentation matrix at time k.

[0129] Step 2.3, the WOA optimization algorithm solves the quadratic programming equation. The real-time control system requires an optimization algorithm with fast convergence speed and high accuracy, and the convergence speed of the WOA algorithm is obviously faster.

[0130] The WOA algorithm is inspired by whale predation, assuming that the current candidate optimal solution or the optimal solution is close to the prey. By defining the best whale search agent position, the search position of other whales updates its own position distance to the best search agent, forming a contraction and circling mechanism, which can be represented by formulas (16) and (17):

[0131] (16).

[0132] (17).

[0133] wherein, denotes the distance between the search position of other whales and the best whale search agent position; denotes the best whale search position vector of the tthiteration, denotes the search position vector of other whales in the search space, i.e. the position vector of other whale individuals in the search space except the best whale position vector found at present; denotes the iteration number; denotes the search position vector of other whales in the search space at the t+1thiteration; denotes the distance vector between the search position of other whales and the best whale search agent position, in the contraction and circling mechanism of the WOA algorithm, is used to calculate the step size and direction of the update of other whale individuals to the best whale position; and Let and denote the position update control coefficient vector and the position update amplitude adjustment coefficient vector, respectively, which can be expressed by equations (18) and (19):

[0134] (18).

[0135] (19).

[0136] where, denotes the iteration attenuation vector, which is linearly reduced from 2 to 0 in the iteration process; denotes the random disturbance vector, which is a random vector in [0, 1]. When denotes that the whale is in the search prey stage, and in the exploration stage, the search agent is randomly selected instead of the best search agent to update its search position, and emphasizes that when allows the algorithm to perform global search, the formula is as follows:

[0137] (20).

[0138] (21).

[0139] where, denotes the position vector of the randomly selected whale search agent at the tth iteration; when is in the whale hunting prey stage, the whale has identified the prey, and equations (20) and (21) are used to update the position.

[0140] In addition to the shrinkage and encirclement mechanism, the simulation of the whale attack also has a spiral attack mode to update the position, and the positions of the whale and the prey can be expressed as:

[0141] (22).

[0142] where, denotes the distance from the whale to the prey in the tth iteration; denotes the distance from the ith whale to the prey, which is also the best solution. is an exponential constant, is the first random number in the interval [-1, 1].

[0143] Assuming that there is a 50% probability of selecting between the shrinkage and encirclement mechanism and the spiral model to update the position of the whale, the mathematics can be expressed as:

[0144] (23).

[0145] is the second random number in the interval [0, 1].

[0146] In summary, the steps of the algorithm can be represented as:

[0147] First initialize the whale population Position. When less than the maximum number of iterations, each whale updates its coefficient , , , and .

[0148] When , the shrinkage mechanism is adopted. If , a search agent is randomly selected and its position is updated, and if , the best search agent position is updated.

[0149] When , the spiral attack mode is adopted, as formula (23).

[0150] Check if there is a search agent beyond the space range, if it is beyond, modify the calculation of the better solution and update , then enter the next iteration until the set number of times is reached.

[0151] The input of the system is obtained by solving the quadratic equation .

[0152] Step 2.4, eliminate noise by introducing Kalman filter. Through the Kalman filter, it can not only play the role of state observer, but also realize the filter function of noise reduction. The discrete state space with noise is represented as:

[0153] (24).

[0154] Where, represents the state vector at the current time k; represents the state transition matrix; represents the state vector at the last time k-1; represents the control input matrix; represents the control input at the last time k-1, the system input calculated by the MPC controller; represents the noise at the last time k-1, such as the wind disturbance existing in the process of landing; represents the observed value of the state vector at the current time k; represents the observation matrix; represents the observation noise at the current time k.

[0155] The prior state estimate is: (25).

[0156] Where, represents the prior state estimate value at the current time k; denotes the posterior state estimate at the previous time instant k-1; denotes the control input at the previous time instant k-1.

[0157] The measurement estimate is: (26).

[0158] wherein, denotes the measurement estimate at the current time instant k; denotes the inverse of the observation matrix. The posterior state estimate is obtained from the two estimate errors:

[0159] (27).

[0160] wherein, denotes the posterior state estimate at the time instant k, the actual value being replaced by the estimate. denotes the Kalman filter gain at the current time instant k. The prior state estimate error and the posterior state estimate error are designed with the Kalman gain as the target:

[0161] (28).

[0162] (29).

[0163] wherein, denotes the prior state estimate error at the current time instant k; denotes the posterior state estimate error at the current time instant k. Substituting equation (24) into equation (29) gives:

[0164] (30).

[0165] The covariance matrix of the prior estimate error is: (31).

[0166] The covariance matrix of the posterior estimate error is:

[0167] (32).

[0168] wherein, denotes the covariance matrix of the prior estimate error at the current time instant k; denotes the expectation of the prior estimate error covariance; denotes the covariance matrix of the posterior estimate error at the current time instant k; denotes the expectation of the posterior estimate error covariance.

[0169] The trace is , representing the variance of the estimate error, ​​denotes the variance of the nth estimated error component at the kth time, find the optimal Kalman gain to make the trace of the posteriori estimation error covariance matrix at the kth time (total error variance) The minimum is acceptable.

[0170] Substitute equations (30) and (31) into equation (32) to obtain:

[0171] (33).

[0172] denotes the covariance matrix of the measurement noise. Let The minimum value is obtained by calculation:

[0173] (34).

[0174] The second derivative is a positive definite matrix, so let (35).

[0175] The Kalman filter gain is obtained: (36).

[0176] According to equation (26), the covariance matrix of the prior state estimation is also needed. Substitute equation (28) into equation (31), because and are independent of each other, so we can obtain:

[0177] (37).

[0178] denotes the expectation of the noise covariance matrix at time k-1; denotes the noise covariance matrix; which is related to Finally, substitute equation (37) into equation (33) and simplify to obtain:

[0179] (38).

[0180] The posteriori state estimation value is obtained by the Kalman filter, which is used as feedback input to replace the predicted state calculated by the MPC controller in equation (8) to update the state of the aircraft.

[0181] Step 2.5, repeat steps 2.1-2.4 until the specified number of cycles.

[0182] Step three, simulation results and analysis.

[0183] ​3.1 Simulation preparation and calculation results

[0184] The reference glide angle of the aircraft = -3.5°, the pitch angle of the aircraft , the glide speed = 70 m / s or so, the height error caused by the influence of deck movement and ship wake turbulence , the angle of attack error and the speed error are expressed as:

[0185] (39).

[0186] (40).

[0187] (41).

[0188] wherein, .

[0189] To avoid excessive calculation, the length of the prediction interval is set to . The number of whale search , the maximum number of iterations . Through simulation, the corresponding error comparison and system input diagram as shown in Figures 3-8 can be obtained.

[0190] 3.2 Result analysis.

[0191] Figure 3 The predicted height error in the simulation is the error of the ITP (ideal touchdown point) position calculated by the MPC, and the a priori height error is the state estimation without the uncertainty term. Both the model predictive control and the a priori state estimation have a large difference from the actual height error, while the posteriori height error estimation after the Kalman filter is almost consistent with the actual height error, which can achieve precise landing. Figure 4 , Figure 5 are the errors of the speed and angle of attack of the aircraft. Due to the influence of external steady-state disturbances, the aircraft state during landing has a certain error compared with the reference value, but it is within the controllable range. Figure 6 and Figure 7 are the changes in the elevator and throttle inputs of the aircraft during landing, which are within the limit of the input.

[0192] The application further provides an application scenario of the automatic carrier landing control method. Specifically, the automatic carrier landing control method provided in the embodiment can be applied in an automatic carrier landing scenario of a carrier-based aircraft. The automatic carrier landing scenario of the carrier-based aircraft includes a pre-landing environment comprehensive evaluation link, an automatic carrier landing precision control link, and a post-landing state review and feedback link. When the carrier-based aircraft returns from a mission and prepares for landing, the carrier-based aircraft enters the pre-landing environment comprehensive evaluation link from a mission airspace. Various sensors on the aircraft carrier and the carrier-based aircraft itself work cooperatively to comprehensively collect information about the motion state of the deck of the aircraft carrier, the disturbance of the wake flow of the aircraft carrier, and the environment such as the weather and the sea state of the landing area, and to preliminarily process and analyze the data to evaluate whether the current environment meets the basic conditions for the carrier-based aircraft to land. If yes, the relevant environmental data are transmitted to the automatic carrier landing precision control link. The automatic carrier landing control method provided in the embodiment belongs to a core part of the automatic carrier landing precision control link. After the data enter the automatic carrier landing precision control link from the pre-landing environment comprehensive evaluation link, the longitudinal dynamics model of the aircraft is discretized to obtain a discrete state space equation, and data such as the longitudinal motion state quantity of the aircraft at the current time are acquired. The model predictive control is converted into a quadratic programming problem, and the optimal control input sequence is obtained by using the whale optimization algorithm. After the Kalman filtering processing, it is determined whether the landing conditions are met. If not, the processing is repeated. The processing result enters the post-landing state review and feedback link. The post-landing state review and feedback link checks the state of the aircraft after landing, and feeds back the data and result of the entire landing process, thereby providing a basis for subsequent optimization. In summary, the technical solution of the application can ensure that the carrier-based aircraft can stably and reliably complete the landing task under complex marine environment and weather conditions, and greatly improves the safety and success rate of the carrier-based aircraft landing.

[0193] Based on the same inventive concept, the embodiment of the application further provides an automatic carrier landing control system for implementing the automatic carrier landing control method described above. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme described in the above method, and therefore the specific limitations in one or more automatic carrier landing control system embodiments provided below can be referred to the limitations of the automatic carrier landing control method described above, which will not be repeated here.

[0194] In one exemplary embodiment, as shown in Figure 8 An automatic carrier landing control system is provided, including:

[0195] A model discretization module 301 is configured to construct a longitudinal dynamics model of an aircraft and to discretize the model to obtain a discrete state space equation for describing the longitudinal motion of the aircraft.

[0196] A state quantity acquisition module 302 is configured to acquire the longitudinal motion state quantity of the aircraft, the steady-state target quantity of the deck motion of the aircraft carrier, and the steady-state disturbance quantity of the wake flow of the aircraft carrier at the current time.

[0197] The control sequence solving module 303 is configured to take the longitudinal motion state quantity of the aircraft at the current time as an initial motion state, solve the discrete state space equation based on the discrete state space equation, and obtain an optimal predicted control input sequence by using a whale optimization algorithm; the control input includes a rudder deflection angle of an elevator, a rudder deflection angle of a flap, and a deflection angle of a throttle lever.

[0198] The Kalman filter module 304 is configured to filter the longitudinal motion state of the aircraft obtained by controlling the aircraft based on the first control input in the optimal predicted control input sequence by using a Kalman filter to obtain a corrected longitudinal motion state of the aircraft.

[0199] The landing judgment module 305 is configured to perform automatic landing if the corrected longitudinal motion state of the aircraft meets a preset landing condition; the preset landing condition includes that a vertical height error between the aircraft and a target landing point is less than a preset height threshold and a speed error is less than a preset speed threshold.

[0200] The feedback control module 306 is configured to return to the step of obtaining the longitudinal motion state quantity of the aircraft at the current time, the motion steady-state target quantity of the aircraft carrier deck, and the steady-state disturbance quantity of the ship stern flow if the corrected longitudinal motion state of the aircraft does not meet the preset landing condition.

[0201] The present application proposes an automatic landing controller based on the whale optimization algorithm of the fusion of the Kalman filter and the model predictive control to suppress the interference of noise and realize real-time control under the influence of the disturbance and other uncertainties in the landing environment. In the range of input limitation, the fusion control method is compared with the model predictive control, and it is found that the KF-MPC method is closer to the actual error and more accurate in tracking the ideal landing point.

[0202] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0203] The technical features of the above embodiments can be combined in any way. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0204] The principles and implementation manners of the present application are described herein by using specific examples, and the above examples are only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will have changes. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. An automatic carrier landing control method, characterized by, The automatic carrier landing control method comprises: a longitudinal dynamics model of the aircraft is constructed and discretized to obtain a discrete state space equation for describing longitudinal motion of the aircraft; obtaining, at a current time, a longitudinal motion state quantity of the aircraft, a steady-state target quantity of carrier deck motion and a steady-state disturbance quantity of a carrier wake flow; the longitudinal motion state quantity of the aircraft at the current time is taken as an initial motion state, and based on the discrete state space equation, a whale optimization algorithm is used for solving to obtain an optimal predicted control input sequence, specifically including: according to the discrete state space equation, a performance index function of a quadratic programming problem of a deviation between a predicted state sequence and a reference state sequence is constructed; a difference between the longitudinal motion state quantity of the aircraft and a steady-state disturbance term and an expected state is taken as an error value, and the longitudinal motion state quantity of the aircraft in the performance index function is replaced to obtain a quadratic programming problem containing a steady-state disturbance; the steady-state disturbance term includes a steady-state disturbance of the steady-state target quantity of the carrier deck motion and a steady-state disturbance of the steady-state disturbance quantity of the carrier wake flow; the whale optimization algorithm is used for solving the quadratic programming problem containing the steady-state disturbance to obtain the optimal predicted control input sequence; the control input includes a rudder deflection angle of an elevator, a rudder deflection angle of a flap and a deflection angle of a throttle lever; the longitudinal motion state of the aircraft obtained by controlling the aircraft based on a first control input in the optimal predicted control input sequence is filtered by a Kalman filter to obtain a corrected longitudinal motion state of the aircraft; if the corrected longitudinal motion state of the aircraft satisfies a preset carrier landing condition, automatic carrier landing is performed; the preset carrier landing condition includes that a vertical height error between the aircraft and a target landing point is less than a preset height threshold and a speed error is less than a preset speed threshold; if the corrected longitudinal motion state of the aircraft does not satisfy the preset carrier landing condition, returning to the step of obtaining, at a current time, a longitudinal motion state quantity of the aircraft, a steady-state target quantity of carrier deck motion and a steady-state disturbance quantity of a carrier wake flow.

2. The automatic carrier landing control method according to claim 1, characterized by, a longitudinal dynamics model of the aircraft is constructed and discretized to obtain a discrete state space equation for describing longitudinal motion of the aircraft, specifically including: the longitudinal dynamics model of the aircraft is constructed by the following formula: ； ； ； wherein, represents the flight acceleration; represents the aircraft mass; represents the engine thrust; represents the angle of attack; represents the flight drag; represents the gravitational acceleration; represents the glide angle; represents the pitch angle; represents the glide angle rate of change; represents the flight speed; represents the lift; represents the pitch acceleration; represents the aircraft body pitch moment of inertia; represents the pitch moment; represents the rate of change of the aircraft vertical height; represents the dynamic pressure; represents the wing area; represents the lift coefficient; represents the zero lift coefficient; represents the angle of attack influence coefficient on the lift coefficient; represents the elevator deflection angle influence coefficient on the lift coefficient; represents the elevator deflection angle; represents the flap deflection angle influence coefficient on the lift coefficient; represents the flap deflection angle; represents the drag coefficient; represents the zero drag coefficient; represents the square term coefficient of the lift coefficient; represents the mean chord of the wing; represents the pitch moment coefficient; represents the zero angle of attack pitch moment coefficient; represents the angle of attack influence coefficient on the pitch moment coefficient; represents the pitch rate; represents the elevator deflection angle influence coefficient on the pitch moment coefficient; represents the flap deflection angle influence coefficient on the pitch moment coefficient; represents the throttle Mach number coefficient; represents the throttle Mach number; represents the engine altitude coefficient; represents the aircraft vertical height; represents the throttle lever deflection angle; Based on the longitudinal dynamics model of the aircraft, the state vector and the control input are selected, and discretization is performed to obtain a discrete state-space equation for describing the longitudinal motion of the aircraft: ； wherein, represents a predicted state vector at k+1 time instant; represents a state transition matrix; represents a control input matrix; represents a state vector at k time instant; represents a control input at k time instant.

3. The automatic carrier landing control method according to claim 1, characterized by, the steady-state target quantity of the carrier deck motion includes a heave displacement of a center of gravity of the carrier and a pitch angle of the carrier deck; the steady-state target quantity of the carrier deck motion is obtained by the following formula: ； ； wherein, represents the pitch motion angle of the aircraft carrier; represents the current time; represents the random disturbance noise of the pitch angle; represents the heave motion amplitude of the center of gravity of the aircraft carrier; represents the random disturbance noise of the heave displacement; represents the motion amplitude of the target landing point; represents the horizontal position of the target landing point on the deck; represents the horizontal position of the center of gravity of the aircraft carrier; the steady-state disturbance quantity of the carrier wake flow includes a steady-state disturbance quantity of a horizontal carrier wake flow and a steady-state disturbance quantity of a vertical carrier wake flow; the steady-state disturbance quantity of the carrier wake flow is obtained by the following formula: ； wherein, represents a steady-state component of the vertical disturbance of the ship wake; represents a steady-state wind component of the steady-state component of the vertical disturbance of the ship wake; represents a periodic disturbance component of the steady-state component of the vertical disturbance of the ship wake; represents a sea-surface-atmosphere-turbulence disturbance component of the steady-state component of the vertical disturbance of the ship wake; represents a steady-state component of the vertical disturbance of the ship wake; represents a steady-state wind component of the steady-state component of the vertical disturbance of the ship wake; represents a periodic disturbance component of the steady-state component of the vertical disturbance of the ship wake; represents a random disturbance component of the steady-state component of the vertical disturbance of the ship wake.

4. The automatic carrier landing control method according to claim 1, characterized by, the whale optimization algorithm is used for solving the quadratic programming problem containing the steady-state disturbance to obtain the optimal predicted control input sequence, specifically including: initializing positions of a whale population and setting an initial value of an iteration number t as 1; wherein a position of each whale individual in the whale population corresponds to a candidate control input; calculating a fitness value of each whale individual in the whale population in the tthiteration; the fitness value is calculated based on the performance index function; determining a current optimal whale individual position according to the fitness value; updating positions of the other whale individuals in the whale population except the current optimal whale individual according to a preset updating strategy based on the position of the current optimal whale individual, the preset updating strategy comprising a shrink-and-circle mechanism updating strategy and a spiral attack mode updating strategy; letting t=t+1, returning to the step of "calculating fitness values of each whale individual in the whale population in the tth iteration" until a maximum iteration number is reached, and outputting a control input sequence corresponding to the position of the current optimal whale individual as an optimal predicted control input sequence.

5. The automatic carrier landing control method according to claim 4, characterized by, updating positions of the other whale individuals in the whale population except the current optimal whale individual according to a preset updating strategy based on the position of the current optimal whale individual, and specifically comprising: determining a position updating control coefficient vector of the current iteration number; determining a scheme for calculating the position of the whale individual according to the position updating control coefficient vector and the generated random number, and updating the current position of the whale individual; if the random number is less than a preset probability threshold value and an absolute value of the position updating control coefficient vector is less than a preset absolute value threshold value, adopting the shrink-and-circle mechanism updating strategy to update the positions of the other whale individuals in the whale population except the current optimal whale individual; if the random number is less than the preset probability threshold value and the absolute value of the position updating control coefficient vector is greater than or equal to the preset absolute value threshold value, adopting a random selection search agent strategy to update the positions of the other whale individuals in the whale population except the current optimal whale individual; if the random number is greater than the preset probability threshold value, adopting the spiral attack mode updating strategy to update the positions of the other whale individuals in the whale population except the current optimal whale individual.

6. An automatic carrier landing control system characterized by, The automatic carrier landing control system applies the automatic carrier landing control method according to any one of claims 1-5, and the automatic carrier landing control system comprises: a model discretization module configured to construct a longitudinal dynamics model of an aircraft and to discretize the longitudinal dynamics model to obtain a discrete state space equation for describing longitudinal motion of the aircraft; a state quantity acquisition module configured to acquire a longitudinal motion state quantity of the aircraft at a current time, a steady-state target quantity of a carrier deck motion, and a steady-state disturbance quantity of a carrier wake flow; a control sequence solving module configured to take the longitudinal motion state quantity of the aircraft at the current time as an initial motion state, to solve, based on the discrete state space equation, an optimal predicted control input sequence by using a whale optimization algorithm, and to take control inputs including a rudder deflection angle of an elevator, a rudder deflection angle of a flap, and a deflection angle of a throttle lever; a Kalman filter module configured to filter a longitudinal motion state of the aircraft obtained by controlling the aircraft based on a first control input in the optimal predicted control input sequence by using a Kalman filter to obtain a corrected longitudinal motion state of the aircraft; a landing judgment module configured to perform automatic landing if the corrected longitudinal motion state of the aircraft satisfies a preset landing condition, the preset landing condition including that a vertical height error between the aircraft and a target landing point is less than a preset height threshold value and a speed error is less than a preset speed threshold value; a feedback control module configured to return to the step of acquiring the longitudinal motion state quantity of the aircraft at the current time, the steady-state target quantity of the carrier deck motion, and the steady-state disturbance quantity of the carrier wake flow if the corrected longitudinal motion state of the aircraft does not satisfy the preset landing condition.

7. A computer device comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that the processor executes the computer program to implement the automatic carrier landing control method of any one of claims 1-5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the automatic carrier landing control method of any one of claims 1-5.

9. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the automatic carrier landing control method of any one of claims 1-5.

Images