Shipborne fixed-wing aircraft landing guidance law design method and system

Abstract

The invention discloses a shipborne fixed-wing aircraft landing guidance law design method and system, and belongs to the field of aircraft guidance law design. The problems that the multi-coordinate system coupling effect of ship motion and aircraft dynamics is not fully decoupled and an existing method is weak in adaptability to non-stable sea conditions are solved. The method comprises the following steps: defining related coordinate systems including a world coordinate system, a landing coordinate system and a ship body coordinate system; calculating an ideal landing point height change value caused by ship deck movement in real time according to the position of the ideal position in the ship body coordinate system; utilizing a KOA-LSTM hybrid forecasting model to predict a motion heave value of a ship deck at a future critical moment; designing a compensation value for the longitudinal position deviation based on the predicted motion heave value, and carrying out dynamic compensation correction on the longitudinal guide instruction; and according to the compensated guide instruction and the real-time position of the aircraft, calculating the transverse and longitudinal deviation of the aircraft relative to the expected landing trajectory to form closed-loop control. The device is mainly used for offshore operation.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of aircraft guidance law design, and particularly relates to a shipborne fixed-wing aircraft landing guidance law design method. BACKGROUND

[0002] Aircrafts play an increasingly important role in the field of sea operations. Due to the limitation of endurance time, aircrafts need to land accurately on the shipborne platform before the fuel is exhausted. The core challenge of the automatic landing process of the shipborne fixed-wing aircraft is to dynamically calculate the lateral and longitudinal deviation of the aircraft relative to the moving deck to generate real-time guidance instructions. At present, the deck of large ships provides physical landing conditions for small and medium-sized fixed-wing aircrafts, but the six-degree-of-freedom motion of the ship caused by complex sea conditions (especially heave, pitch and roll) will cause the ideal landing point position to continuously deviate, and the traditional guidance method is difficult to meet the high-precision and low-delay control requirements.

[0003] In the prior art, the static baseline method (such as the newly proposed "Analysis of Carrier Landing Safety Based on Baseline Statistical Method") constructs a standard operation baseline through historical data to quantify the operation deviation. However, this method relies on a static statistical model and cannot effectively respond to sudden wind, sudden sea wave and other unmodeled disturbances, which easily introduces evaluation deviation in the dynamic environment, resulting in a mismatch between the guidance instructions and the actual motion. The predictive control method (such as the "Landing Control Technology Based on Self-correcting MPC" proposed by Han Wei) uses an AR model to predict the deck motion and combines the least squares identification to update the aerodynamic derivatives in real time. Although the model accuracy is improved, there are real-time and model dependency problems, and the method for calculating the deviation is not provided.

[0004] The more essential bottleneck is that the multi-coordinate system coupling effect of ship motion and aircraft dynamics has not been fully decoupled. The traditional method usually ignores the instantaneous height change of the deck motion calculation or uses a simplified linear model to compensate, and the error is amplified sharply in the severe shaking condition. In addition, the existing prediction model (such as AR, Kalman filter) has weak adaptability to non-stationary sea conditions, and the root mean square error (RMSE) is generally more than 15%, resulting in longitudinal compensation lag or over-regulation. SUMMARY

[0005] Therefore, the present application aims to provide a shipborne fixed-wing aircraft landing guidance law design method and system to solve the problems of the multi-coordinate system coupling effect of ship motion and aircraft dynamics not being fully decoupled and the weak adaptability of the existing method to non-stationary sea conditions.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solutions: A shipborne fixed-wing aircraft landing guidance law design method, the method comprising: Step 1: defining related coordinate systems, including: a world coordinate system, a landing coordinate system, and a ship body coordinate system; Step 2: Real-time calculation of the ideal landing point height change value caused by the ship deck movement according to the position in the ship body coordinate system; Step 3: Predict the ship deck movement heave value at the future key moment by using the KOA-LSTM hybrid prediction model; Step 4: Based on the predicted movement heave value, design the compensation value of the longitudinal position deviation, and dynamically compensate and correct the longitudinal guidance instruction; Step 5: According to the compensated guidance instruction and the real-time position of the aircraft, calculate the lateral and longitudinal deviation of the aircraft relative to the expected landing trajectory, and form a closed-loop control.

[0007] Further, an optimal mode is also proposed, wherein the calculation of the ideal landing point height change value caused by the ship deck movement in step 2 comprises: Designing an ideal landing point conversion function from the body ship coordinate system to the world coordinate system:

[0008] Wherein, is the angle between the landing coordinate system and the ship body coordinate system, , is a rotation transformation matrix, is the aircraft yaw angle, is the aircraft pitch angle, is the aircraft roll angle, is the position coordinate of the ideal landing point in the ship body coordinate system, is the position coordinate of the ship in the world coordinate system; Calculate the static deck ideal landing point position and dynamic deck ideal landing point coordinate, and get the height change value through the formula, wherein, represents the ideal landing point in the landing coordinate system under the consideration of ship deck lateral and longitudinal rolling and vertical oscillation, axis coordinate, represents the ideal landing point in the landing coordinate system under the consideration of deck lateral and longitudinal rolling and vertical oscillation, axis coordinate.

[0009] Further, an optimal mode is also proposed, wherein the KOA-LSTM hybrid prediction model comprises: Initializing the parameters of KOA, including: the number of planets N, the maximum number of iterations Tmax, the upper and lower bounds of hyperparameters, and the control parameters , , ; Initialize the celestial body population with random position, orbital eccentricity and orbital period, and evaluate the fitness value of the initial population. The root mean square error is used as the fitness function to train the model; Mark the global optimal solution as the position of the sun; Calculate the Euclidean distance between the sun and all celestial bodies, the gravitational force, and the velocity of all celestial bodies; Update the distance between the object and the sun or the new position of the object; Calculate the fitness value of all celestial bodies and the sun, and update the global optimal solution until the maximum number of iterations is reached.

[0010] Further, a preferred mode is also proposed, and the prediction of the future key moment of the motion heave value comprises: Collect the roll, pitch, and heave motion posture data of the ship every 0.05 seconds; Convert the ship motion posture into a time series input-output pair, and use the data of the past 50 time steps to predict the ship posture in the future 5 seconds; According to the ratio of 6:2:2, the data of the three motion postures are respectively divided into training set, test set and validation set, and standardized processing is performed; The standardized training set is input into the KOA-LSTM hybrid prediction model, and the predicted motion heave value is output.

[0011] Further, a preferred mode is also proposed, and the dynamic compensation correction in step 4 comprises: Design a compensation amount transfer function: , , , , wherein, represents a second-order compensator transfer function, represents a first-order integrator transfer function, represents a lead-lag compensator transfer function, represents a second-order system transfer function with zero point, represents the Laplace operator, =1, =0.8, =1.2, =1, =1.5, =1, =1, =1, =0.5, =0.1, =1, =0.3, =1, =0.5, =1, =1, =1; Set the deck motion estimation threshold to ; Calculate the compensation value of deck motion for longitudinal deviation ;

[0012] in, is the predicted deck heave value in the future. It is the real-time deck heave value.

[0013] Furthermore, a preferred method is proposed, wherein the calculation of the lateral and longitudinal deviations of the aircraft relative to the desired landing trajectory in step 5 includes: ignoring the deck pitch and roll conditions, and calculating the aircraft's geodetic coordinate system position. Convert to the landing coordinate system and get the coordinates ; in, The world coordinate system of the aircraft Axis position, The world coordinate system of the aircraft Axis position, The world coordinate system of the aircraft Axis position, The aircraft is in the landing coordinate system Axis position, The aircraft is in the landing coordinate system Axis position, The aircraft is in the landing coordinate system Axis position; Calculate the ideal vertical glide position value: , in, Indicates the ideal glideslope angle of the aircraft; Calculate the height deviation without taking deck motion compensation into account:

[0014] Calculate the height deviation value for dynamic compensation taking into account deck motion compensation:

[0015] in, The distance between the fixed-wing aircraft and the ideal landing point Axis distance.

[0016] Furthermore, a preferred method is proposed, wherein the step 1 of defining the relevant coordinate system includes: World coordinate system origin The sea level at the initial position of the ship is defined by the North-East direction, Pointing to the North Pole of the Earth, Pointing to the East direction, The direction is consistent with the direction of gravity.

[0017] The origin of the landing coordinate system It is the ideal landing point on the deck of the ship when the aircraft lands, The axis is horizontally pointing to a specified angle of the deck of the ship, The axis is perpendicular to the deck surface and faces downward, The axis is determined according to the right-hand rule; The origin of the ship body coordinate system Fixed at the intersection of the waterline and the centerboard surface of the ship, located at the bottom of the bow column, The axis points to the bow direction along the centerboard surface, The axis is perpendicular to the designed waterline, The axis points to the right side according to the right-hand rule.

[0018] Based on the same inventive concept, the present application also provides a shipborne fixed-wing aircraft landing guidance law design system, the system comprising: A coordinate system definition unit for defining relevant coordinate systems, including: a world coordinate system, a landing coordinate system, and a ship body coordinate system; A height change value calculation unit for calculating the height change value of the ideal landing point caused by the movement of the ship deck in real time according to the position of the ideal position in the ship body coordinate system; A prediction unit for predicting the motion heave value of the ship deck at a future key moment by using a KOA-LSTM hybrid prediction model; A compensation unit for designing a compensation value for the longitudinal position deviation based on the predicted motion heave value, and dynamically compensating and correcting the longitudinal guidance instruction; A closed-loop control unit for calculating the lateral and longitudinal deviation of the aircraft relative to the expected landing trajectory according to the compensated guidance instruction and the real-time position of the aircraft, and forming a closed-loop control.

[0019] Based on the same inventive concept, the present application also provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the shipborne fixed-wing aircraft landing guidance law design method according to any one of the above.

[0020] Based on the same inventive concept, the present application also provides a computer readable storage medium, which stores a computer program, and when the computer program is run by a processor, the steps of the shipborne fixed-wing aircraft landing guidance law design method according to any one of the above are executed.

[0021] Compared with the prior art, the present application has the beneficial effects that: The method provided by the present application solves the prediction drift problem of the traditional method (such as the AR model) based on the KOA-LSTM hybrid prediction model and the dynamic compensation strategy. Specifically, the KOA-LSTM model dynamically optimizes the LSTM network hyperparameters through the Kepler optimization algorithm (KOA), takes the root mean square error (RMSE) as the fitness function, and realizes accurate prediction of the future 5-second deck heave motion. Compared with the traditional AR model, the prediction time domain is extended to 5 seconds, and the root mean square error is reduced by more than 30%, effectively avoiding the drift phenomenon in long-time domain prediction. Combined with the compensation transfer function and the threshold mechanism, the system can prospectively correct the longitudinal guidance command, so that the landing height error is stabilized within a small range (for example, within ±0.5 meters) under the condition of severe deck shaking. This effect is verified by virtual simulation, and the prediction accuracy is significantly improved.

[0022] Further, the deck heave motion prediction curve shows that the error rate of the present application is only about 7% under complex sea conditions, which confirms its high-precision characteristics. It combines machine learning optimization and physical model compensation, breaks through the limitations of existing technologies that rely on historical data or simplified linear models, and thus improves the landing success rate by more than 10%.

[0023] The present application solves the problem of real-time performance in the prediction control method through an innovative calculation optimization mechanism. Specifically, it includes a compensation amount threshold truncation mechanism that only uses the predicted value, otherwise switches to the measured value, and a lateral deviation static treatment that ignores deck motion coupling and directly calculates. These mechanisms reduce invalid optimization iterations and real-time calculation complexity, reducing single-step calculation time to milliseconds (for example, <10 milliseconds), meeting the strict control cycle requirements (usually 20-50 milliseconds) of shipborne fixed-wing aircraft. Compared with Hanwei's self-correcting MPC method (which relies on high-complexity model identification), the present application avoids the computational burden of real-time calculation of motion coupling, achieving more than 50% delay optimization. In the event of a sudden wind or sea wave, the system can quickly respond to ensure timely output of the guidance command.

[0024] The present application fuses multi-coordinate system motion decoupling (such as the definition and conversion of the world coordinate system, the landing coordinate system, and the ship body coordinate system), the system can explicitly calculate the ideal landing point height variation, and effectively offset the influence of the ship's six degrees of freedom motion in combination with dynamic compensation. This makes the guidance law have strong robustness in all sea conditions: simulation verification ( Figures 5-7 ) shows that under the condition of severe deck heave ( Figure 5 ), the lateral and longitudinal deviations are stably controlled within the allowable range: the lateral deviation curve ( Figure 6 ) shows that static treatment makes the fluctuation decrease and the error decrease by 40%: the longitudinal deviation curve (Figure 7 ) verified that after dynamic compensation converge faster, and the landing safety is improved by more than 20%, which breaks through the vulnerability of the original new re-static baseline method to unmodeled disturbances, and realizes self-adaptation in all working conditions. BRIEF DESCRIPTION OF DRAWINGS

[0025] The accompanying drawings, which form a part of this application, are included to provide a further understanding of the application, illustrate embodiments of the application and assist in the explanation of the application. The drawings included are: Figure 1 A flow chart of a shipborne fixed-wing aircraft landing guidance law design method according to the present application; Figure 2 A schematic diagram of the related coordinate system according to the present application, wherein, is the angle between the deck and the horizontal plane of the ship; Figure 3 A flow chart of the KOA-LSTM prediction method according to the present application; Figure 4 A schematic diagram of the deck heave motion prediction curve according to the present application; Figure 5 A schematic diagram of the ship deck heave curve according to the present application; Figure 6 A schematic diagram of the aircraft lateral deviation curve according to the present application; Figure 7 A schematic diagram of the aircraft longitudinal deviation curve according to the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict, and the described embodiments are only part of the embodiments of the present application, not all the embodiments.

[0027] Embodiment one, see Figure 1 This embodiment is described. The shipborne fixed-wing aircraft landing guidance law design method described in this embodiment comprises: Step 1: define the related coordinate system, including: world coordinate system, landing coordinate system, ship body coordinate system; Step 2: calculate the ideal landing point height change value caused by the ship deck motion in real time according to the position of the ideal position in the ship body coordinate system; Step 3: predict the heave value of the ship deck at the key moment in the future by using the KOA-LSTM hybrid prediction model; Step 4: Based on the predicted motion heave value, a compensation value for the longitudinal position deviation is designed to dynamically compensate and correct the longitudinal guidance instruction; Step 5: According to the compensated guidance instruction and the real-time position of the aircraft, the lateral and longitudinal deviations of the aircraft relative to the expected landing trajectory are calculated to form a closed-loop control.

[0028] The flow chart of the embodiment is shown in Figure 1 As shown in the figure, in this embodiment, the relevant coordinate system required for calculation is first defined as the coordinate system basis for subsequent steps, and an algorithm for real-time calculation of the instantaneous change value of the ideal landing point height caused by the ship deck motion is designed to obtain the vertical displacement of the ideal landing point. On this basis, the current and historical motion data are used to predict the motion attitude value (including heave, pitch and roll) of the ship deck at the future key time (such as the predicted landing time) through the prediction algorithm, and the longitudinal guidance instruction is dynamically compensated and corrected in advance according to these predicted attitude values to offset the influence of the predicted deck motion on the future landing point position. Based on the guidance instruction after compensation and correction and the real-time position of the aircraft, the real-time deviations of the aircraft in the lateral and longitudinal directions relative to the expected landing trajectory are calculated to form a closed-loop control system or provide precise guidance information for the pilot, thereby achieving more accurate and stable landing.

[0029] The method proposed in this embodiment is a dynamic guidance law design method for shipborne fixed-wing aircraft. Through the innovative dynamic compensation framework and efficient prediction mechanism, the following technical effects are achieved: (1) The guidance accuracy is significantly improved: based on the KOA-LSTM hybrid prediction model (prediction time domain of 5 seconds), the root mean square error of deck heave motion prediction is reduced by more than 30%, overcoming the long-time domain prediction drift problem of traditional AR model. And through the compensation strategy, the landing height error is stabilized within a small range under the condition of severe deck shaking. (2) Real-time optimization is broken through: the compensation amount threshold truncation mechanism is adopted to reduce invalid optimization calculation; the lateral deviation is statically processed to avoid real-time calculation of motion coupling, the single-step calculation time is reduced, and the strict requirements of shipborne fixed-wing aircraft landing control period are met. This method finally realizes high-precision, low-latency and strong-robust shipborne fixed-wing aircraft landing guidance.

[0030] Embodiment two, this embodiment is a further limitation of the shipborne fixed-wing aircraft landing guidance law design method described in embodiment one, the step 2 of calculating the ideal landing point height change value caused by the ship deck motion includes: The ideal landing point is designed to be converted from the body ship coordinate system to the world coordinate system function:

[0031] Wherein, is the angle between the landing coordinate system and the ship body coordinate system, , is a rotation transformation matrix, is the aircraft yaw angle, is the aircraft pitch angle, is the aircraft roll angle, is the position coordinate of the ideal landing point in the ship body coordinate system, is the position coordinate of the ship in the world coordinate system; Calculate the static deck ideal landing point position and dynamic deck ideal landing point coordinate, and get the height change value by formula , where, represents the axis coordinate of the ideal landing point in the landing coordinate system considering the ship deck roll, pitch and heave, represents the axis coordinate of the ideal landing point in the landing coordinate system without considering the deck roll, pitch and heave.

[0032] The height change value of the ideal landing point caused by the ship deck motion can effectively consider the influence of the ship motion on the aircraft landing. This is crucial for the landing of shipborne fixed-wing aircraft in dynamic environment, because the ship's wave, roll, pitch and other motions will affect the landing position and landing accuracy of the aircraft.

[0033] By rotating the transformation matrix to convert the landing coordinate system and the ship body coordinate system, more accurate landing guidance design can be achieved. By considering the attitude change of the aircraft through angles (aircraft yaw angle, pitch angle, roll angle), the interference of attitude change on landing can be effectively reduced, and the safety and accuracy of landing can be improved.

[0034] In complex dynamic environment, by accurately considering the motion change of the ship, the aircraft can be closer to the expected target position during landing, and the error caused by environmental change can be reduced, thereby improving the landing accuracy.

[0035] Embodiment three, the embodiment is a further limitation of the shipborne fixed-wing aircraft landing guidance law design method of embodiment one, the KOA-LSTM hybrid prediction model comprises: Initialize the parameters of KOA, including: number of planets N, maximum number of iterations Tmax, upper and lower bounds of hyperparameters, control parameters , , ; Initialize the population of celestial bodies with random positions, orbital eccentricity and orbital period, and evaluate the fitness value of the initial population. The root mean square error is used as the fitness function to train the model; Mark the global optimal solution as the position of the sun; Calculate the Euclidean distance between the sun and all celestial bodies, the gravitational force, and the velocity of all celestial bodies; Update the distance between the object and the sun or the new position of the object; Calculate the fitness value of all celestial bodies and the sun, and update the global optimal solution until the maximum number of iterations is reached.

[0036] Embodiment four, this embodiment is a further limitation of the shipboard fixed-wing aircraft landing guidance law design method described in embodiment one, the prediction of the future key moment motion heave value includes: Collect the roll, pitch, and heave motion attitude data of the ship every 0.05 seconds; Convert the ship motion attitude into a time series input-output pair, and use the data of the past 50 time steps to predict the ship attitude in the future 5 seconds; Divide the data of the three motion attitudes into training set, test set, and validation set in the ratio of 6:2:2 respectively, and perform standardization processing; Input the standardized training set into the KOA-LSTM hybrid prediction model, and output the predicted motion heave value.

[0037] This embodiment is explained in combination with embodiment three. This embodiment uses a hybrid model of KOA (Kermans optimization algorithm) combined with LSTM (long short-term memory network), which considers both nonlinear factors and time series regularity. LSTM effectively remembers and processes historical ship motion information through its gating mechanism, while KOA optimizes the model through global search, enhancing the robustness and adaptability of the model and enabling it to cope with complex ship motion patterns and environmental changes. By collecting the roll, pitch, and heave motion attitude data of the ship every 0.05 seconds, more detailed motion trajectory information can be obtained. Using these high-frequency data to train the model helps improve the prediction accuracy of the ship's attitude changes at future key moments. The data of the three motion attitudes are divided into training set, test set, and validation set in the ratio of 6:2:2 respectively, and standardized processing is performed. This data processing method helps to eliminate the dimensional differences between data, improve the training efficiency and prediction effect of the model. At the same time, reasonable division of the training set can effectively avoid overfitting problem and enhance the generalization ability of the model.

[0038] During each iteration of the model, as the fitness value of the celestial bodies is updated, the prediction result can be optimized in real time. In ship motion prediction, the model can continuously adjust the prediction with each data input, respond to changes in ship attitude in a timely manner, and ensure the accuracy and timeliness of the prediction result at future key moments.

[0039] The method provided by the embodiment is based on a KOA-LSTM hybrid prediction model and a dynamic compensation strategy, and solves the prediction drift problem of a traditional method (such as an AR model). Specifically, the KOA-LSTM model dynamically optimizes the LSTM network hyperparameters through a Kepler optimization algorithm (KOA), takes a root mean square error (RMSE) as a fitness function, and realizes accurate prediction of deck heave motion in the next 5 seconds. Compared with the traditional AR model, the prediction time domain is extended to 5 seconds, and the root mean square error is reduced by more than 30%, effectively avoiding the drift phenomenon in long-time domain prediction. In combination with a compensation transfer function and a threshold mechanism, the system can prospectively correct the longitudinal guidance instruction, so that the landing height error is stabilized in a small range (for example, within ±0.5 meters) under the condition of severe deck shaking. This effect is verified by virtual simulation, and the prediction accuracy is significantly improved.

[0040] Embodiment five, this embodiment is a further limitation of the shipboard fixed-wing aircraft landing guidance law design method described in embodiment one, the dynamic compensation correction in step 4 includes: Design the compensation amount transfer function: , , , , Among them, represents a second-order compensator transfer function, represents a first-order integrator transfer function, represents a lead-lag compensator transfer function, represents a two-order system transfer function with zero point, represents the Laplace operator, =1, =0.8, =1.2, =1, =1.5, =1, =1, =1, =0.5, =0.1, =1, =0.3, =1, =0.5, =1, =1, =1; Set the deck motion estimation threshold to ; Calculate the compensation value of the deck motion to the longitudinal deviation ;

[0041] wherein, is the predicted heave value of the deck for a future period of time, is the real-time heave value of the deck.

[0042] Embodiment six, this embodiment is a further limitation of the method for designing a landing guidance law for a seaborne fixed-wing aircraft according to embodiment one, wherein the step 5 of calculating the lateral and longitudinal deviation of the aircraft relative to the desired landing trajectory comprises: under the condition of ignoring the deck roll, pitch and heave, converting the aircraft geodetic coordinate system position to the landing coordinate system to obtain the coordinates ; wherein, is the position of the aircraft in the world coordinate system axis direction, is the position of the aircraft in the world coordinate system axis direction, is the position of the aircraft in the world coordinate system axis direction, is the position of the aircraft in the landing coordinate system axis direction, is the position of the aircraft in the landing coordinate system axis direction, is the position of the aircraft in the landing coordinate system axis direction; calculating the ideal vertical glide position value: , wherein, represents the ideal glide angle value of the aircraft; calculating the height deviation without considering deck motion compensation:

[0043] calculating the height deviation value considering deck motion compensation for dynamic compensation:

[0044] wherein, is the set fixed-wing aircraft distance from the ideal landing point axis direction.

[0045] In this embodiment, the calculation process is more concise by ignoring the influence of deck roll, pitch and heave. Although the actual deck movement has a certain influence on the landing process of the aircraft, this simplification can ensure sufficient accuracy for most landing scenarios. By converting the aircraft's geodetic coordinate system position to the landing coordinate system, the relative position of the aircraft can be clearly and accurately obtained, avoiding conversion errors between coordinate systems and improving accuracy and stability. Calculating the ideal vertical glide position value of the aircraft and calculating based on the glide angle can more accurately control the change in aircraft altitude and provide a basis for subsequent dynamic compensation. By considering the dynamic compensation of deck movement, the influence of deck movement on aircraft altitude can be effectively reduced, ensuring that the aircraft can stably maintain the expected flight trajectory during landing, thereby improving the safety and reliability of the aircraft landing process.

[0046] In combination with the above steps, especially during dynamic compensation, by real-time correction of the height deviation of the aircraft, it helps to cope with the instability of the ship deck, thereby improving the landing accuracy of the aircraft in complex environments and reducing possible deviations during landing.

[0047] Embodiment seven, this embodiment is a further limitation of the method for designing a shipborne fixed-wing aircraft landing guidance law described in embodiment one, wherein the step 1 of defining the related coordinate systems comprises: the origin of the world coordinate system the sea level at the initial position of the ship, defined by the North East Earth, pointing to the Earth's North Pole, pointing to the east direction, the direction is consistent with the direction of gravity.

[0048] the origin of the landing coordinate system is the ideal landing point on the deck of the ship when the aircraft lands, the axis is horizontally pointing to the designated angle of the deck of the ship, the axis is perpendicular to the deck surface and points downward, the axis is determined according to the right-hand rule; the origin of the ship body coordinate system is fixed to the intersection of the waterline and the centerboard surface of the ship, located at the bottom of the bow column, the axis points to the bow direction along the centerboard surface, the axis is perpendicular to the designed waterline, the axis points to the right side according to the right-hand rule.

[0049] Embodiment eight, a system for designing a shipborne fixed-wing aircraft landing guidance law, the system comprises: a coordinate system definition unit for defining related coordinate systems, including: a world coordinate system, a landing coordinate system, and a ship body coordinate system; a height change value calculation unit configured to calculate a height change value of the ideal landing point caused by the ship deck movement in real time according to the position of the ideal position in the ship body coordinate system; a prediction unit configured to predict the ship deck movement heave value at a future key moment by using the KOA-LSTM hybrid prediction model; a compensation unit configured to design a compensation value for the longitudinal position deviation based on the predicted movement heave value, and to dynamically compensate and correct the longitudinal guidance instruction; a closed-loop control unit configured to calculate the lateral and longitudinal deviation of the aircraft relative to the expected landing trajectory according to the compensated guidance instruction and the real-time position of the aircraft, and to form a closed-loop control.

[0050] Embodiment nine, a computer device according to the embodiment, comprising a memory and a processor, the memory stores a computer program, when the processor runs the computer program stored in the memory, the processor executes the shipborne fixed-wing aircraft landing guidance law design method according to any one of embodiments one to seven.

[0051] Embodiment ten, a computer readable storage medium according to the embodiment, the computer readable storage medium stores a computer program, when the computer program is run by a processor, the steps of the shipborne fixed-wing aircraft landing guidance law design method according to any one of embodiments one to seven are executed.

[0052] Embodiment eleven, see Figures 2 to 7 This embodiment is described. This embodiment is a specific embodiment of the shipborne fixed-wing aircraft landing guidance law design method according to embodiment one, and is also used to explain embodiments two to seven, specifically: Step 1: define the relevant coordinate system required for calculation, as the coordinate system basis for the subsequent steps, such as Figure 2 As shown in the figure, the relevant coordinate system includes: World coordinate system: the origin of the world coordinate system The sea level at the initial position of the ship is defined by the North East Earth, wherein Points to the north pole of the earth, Points to the east direction, The direction is consistent with the direction of gravity, as Figure 2 As shown in the left figure; Landing coordinate system: the origin of the landing coordinate system Is the ideal landing point on the ship deck when the aircraft lands, The axis is horizontally directed to the designated angle of the ship deck, and it is emphasized here that the direction may not be consistent with the direction of the bow of the ship, The axis is perpendicular to the deck and points downward, The axis is consistent with the plane Satisfy the right-hand rule, such as Figure 2 As shown in the right figure; Ship body coordinate system: the origin of the ship body coordinate system Fixed at the intersection of the ship waterline and the centerboard, located at the bottom of the bow column, The axis points to the bow direction along the centerboard, and the positive direction is the forward direction of the ship, The axis is perpendicular to the design waterline, and the positive direction is perpendicular upward, The axis is determined by the right-hand rule, and the positive direction points to the right side.

[0053] Step 2: Calculate the ideal drop point height change value caused by the ship deck movement, the specific implementation steps are as follows: Step 21: Define the ideal position in the ship body coordinate system: ; Step 22: Design the ideal drop point conversion function from the body ship coordinate system to the world coordinate system, the specific form is as follows: (1) In the above formula: is the drop coordinate system and the angle between the ship body coordinate system , , is the rotation transformation matrix, respectively for the aircraft yaw angle, pitch angle, roll angle; Step 23: Calculate the static deck ideal drop point position, that is, without considering the deck pitch, heave and roll, define , , , is the height value of the deck from the sea surface in the static state, which is a constant, and the coordinates of the ideal drop point in the drop coordinate system without considering the deck pitch, heave and roll are calculated by formula (1) ; Step 24: Calculate the dynamic deck ideal drop point position, that is, considering the deck pitch, heave and roll, the coordinates of the ideal drop point in the drop coordinate system considering the deck pitch, heave and roll are calculated by formula (1) ; Step 25: Calculate the ideal drop point height change value caused by the ship deck movement using the following formula: (2) Step 3: Build a KOA-LSTM attitude prediction model to calculate the future motion heave value of the ship deck; In this embodiment, the motion characteristics of ships and shipboard fixed-wing aircraft are different, especially in the frequency domain, the response characteristics are quite different. Therefore, the related algorithm is designed to predict the future motion posture value of the ship deck, and the specific steps are as follows: Step 31: Collect the roll, pitch and heave motion posture data of the actual ship every 0.05 seconds, a total of 20000 data points, i.e. 1000 seconds of data.

[0054] Step 32: Convert the ship motion posture into input-output data pairs with clear time correlation, which is a supervised learning format suitable for LSTM neural network training. The task to be completed is to use the data of the past 50 time steps to predict the ship posture at the 5th second in the future, so each sample needs to include posture data of the past 50 time steps and the future 100 time steps.

[0055] Step 33: Divide the data of the three motion postures into training set, test set and validation set according to the ratio of 6:2:2; then standardize the processing to eliminate the dimension effect, and finally reconstruct the data into the three-dimensional structure required by the LSTM network [sample number, time step, feature number]; Step 34: Construct the KOA-LSTM posture prediction model, including: Step 341: Prepare the ship posture data that meets the above requirements. Use the training set and the validation set to train and validate the prediction model, and use the test set to test the KOA-LSTM model; Step 342: Initialize the parameters of KOA: number of planets N, maximum number of iterations Tmax, upper and lower bounds of hyperparameters (learning rate, number of neurons), control parameters 、 、 ; Step 343: Initialize the celestial body population with random position, orbital eccentricity and orbital period, and evaluate the fitness value of the initial population. The root mean square error (RMSE) is used as the fitness function to train the model; Step 344: Mark the global optimal solution as the position of the sun, i.e. the current hyperparameter selection; Step 345: Calculate the Euclidean distance between the sun and all celestial bodies, the gravity, and the speed of all celestial bodies; Step 346: Update the distance between the object and the sun or the new position of the object; Step 347: Calculate the fitness value of all celestial bodies and the sun, and update the global optimal solution; Step 348: Save the optimal solution after iteration, which is the optimal parameter selection of the LSTM network; When predicting the time series of ship posture, the selection of hyperparameters plays a decisive role in the prediction accuracy of the LSTM network. According to the above method, the heave value of the deck in the future can be obtained. .

[0056] Step 4: After considering the future heave of the ship's deck, design the compensation value for the longitudinal position deviation. The specific steps are as follows: Step 41: Design the compensation transfer function. , , , , in the above formula and These are undetermined parameters. In this embodiment, simulation is used to test and determine the above parameters: =1, =0.8, =1.2, =1, =1.5, =1, =1, =1, =0.5, =0.1, =1, =0.3, =1, =0.5, =1, =1, =1; Step 42: Design the deck motion estimation validity and set the deck motion estimation threshold to , this value can be set according to the actual situation. In this implementation, =3.5m; Step 43: Calculate the compensation value of deck motion for longitudinal deviation according to the following strategy : (3) Step 5: Calculate the lateral and longitudinal deviations of the aircraft relative to the ideal glide path, using the following steps: Step 51: Considering the robustness of the tracking performance of the shipborne fixed-wing aircraft, the longitudinal deviation of the aircraft is calculated by ignoring the deck pitch and heave conditions, that is, defining , , ,in is the height of the deck from the sea surface in the static state, which is a constant. Let the coordinates of the aircraft in the geodetic coordinate system be , the coordinates of the aircraft in the landing coordinate system can be calculated by formula (1) without considering the deck pitch and roll ; Step 52: Calculate the ideal vertical sliding position value: ,in Indicates the ideal glideslope angle of the aircraft; Step 53: Calculate the height deviation without considering deck motion compensation: ; Step 54: Calculate the aircraft's lateral deviation: ; Step 55: Calculate the height deviation value when considering deck motion compensation, as shown in the following formula: (4) The present invention carries out the simulation effect of virtual scenes based on the above method, which is specifically divided into two scenes: deck motion prediction effect simulation and aircraft lateral and longitudinal deviation calculation effect simulation The present invention aims at the deck motion prediction function, and uses historical data to predict the deck heave value in the next 5 seconds. The curve is as follows: Figure 4 As shown, the deck heave prediction error in this invention is approximately 7%, demonstrating excellent results. This demonstrates its high precision. By integrating machine learning optimization with physical model compensation, this method overcomes the limitations of existing technologies that rely on historical data or simplified linear models, thereby increasing landing success rates by over 10%.

[0057] The present invention integrates multi-coordinate system motion decoupling (such as the definition and conversion of the world coordinate system, landing coordinate system and ship body coordinate system). The system can explicitly calculate the change in the ideal landing point height and, combined with dynamic compensation, effectively offset the impact of the ship's six-degree-of-freedom motion. This makes the guidance law highly robust in all sea conditions: Figures 5 to 7 As shown, the method proposed in the present invention can calculate the lateral and longitudinal deviations of the aircraft during landing on the ship deck in real time according to the ship's motion state and the aircraft's motion state, that is, to achieve the design task of the guidance law of the ship-borne fixed-wing aircraft under severe deck heave conditions ( Figure 5 ), the lateral and longitudinal deviations are stably controlled within the allowable range: lateral deviation curve ( Figure 6 ) shows that static treatment reduces fluctuations and errors by 40%: longitudinal deviation curve ( Figure 7 ) confirmed that the convergence is faster after dynamic compensation, improving landing safety by more than 20%. It breaks through the vulnerability of the original new static baseline method to unmodeled disturbances and achieves full working condition adaptation.

[0058] Those skilled in the art will appreciate that embodiments of the disclosure can be supplied as a method, a system, or a computer program product. Thus, the disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the disclosure can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer-readable program code.

[0059] The disclosure is described in reference to the flowchart and / or block diagrams of the method, apparatus (system) and computer program product according to embodiments of the disclosure. It should be understood that each flow and / or block in the flowchart and / or block diagram and a combination of flows and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the flow Figure 1 one or more flows and / or blocks Figure 1 one or more flows and / or blocks Figure 1 one or more flows and / or blocks Figure 1 one or more flows and / or blocks one or more flows and / or blocks Figure 1 one or more flows and / or blocks Figure 1 one or more flows and / or blocks Finally, it should be noted that the above embodiments are merely used to illustrate the technical solutions of the present disclosure, rather than limiting the scope of protection of the present disclosure. Although the present disclosure has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading the present disclosure, they can make various changes, modifications or equivalent replacements to the specific embodiments of the present disclosure. However, these changes, modifications or equivalent replacements are all within the scope of protection of the disclosed patent claims.

Claims

1. A method for designing a landing guidance law for a shipborne fixed-wing aircraft, characterized in that: The method comprises: Step 1: Define relevant coordinate systems, including: world coordinate system, landing coordinate system, and ship body coordinate system; Step 2: Calculate the height change of the ideal landing point caused by the ship deck movement in real time based on the position of the ideal position in the ship's coordinate system; Step 3: Use the KOA-LSTM hybrid prediction model to predict the ship deck's heave and sink values ​​at key moments in the future; Step 4: Based on the predicted motion heave value, design the compensation value for the longitudinal position deviation and dynamically compensate and correct the longitudinal guidance command; Step 5: Based on the compensated guidance instructions and the real-time position of the aircraft, the lateral and longitudinal deviations of the aircraft relative to the desired landing trajectory are calculated to form a closed-loop control.

2. The method for designing a landing guidance law for a shipborne fixed-wing aircraft according to claim 1, characterized in that: The calculation of the ideal landing point height change value caused by the ship deck movement in step 2 includes: Design the ideal landing point by converting the coordinate system of the ship to the world coordinate system: in, is the angle between the landing coordinate system and the ship's coordinate system, , is the rotation transformation matrix, is the aircraft yaw angle, is the aircraft pitch angle, is the aircraft roll angle, is the coordinate of the ideal landing point in the ship's coordinate system, is the position coordinate of the ship in the world coordinate system; Calculate the ideal landing point position of the static deck and the ideal landing point coordinates of the dynamic deck, and use the formula Get the height change value, where It represents the ideal landing point in the landing coordinate system considering the ship deck's pitch and roll. Axis coordinates, Indicates the ideal landing point in the landing coordinate system without considering the deck's pitch and roll Axis coordinates.

3. The method for designing a landing guidance law for a shipborne fixed-wing aircraft according to claim 1, characterized in that: The KOA-LSTM hybrid forecasting model includes: Initialize KOA parameters, including: number of planets N, maximum number of iterations Tmax, upper and lower bounds of hyperparameters, and control parameters 、 、 ; Initialize the celestial body population with random positions, orbital eccentricities and orbital periods, evaluate the fitness value of the initial population, and use the root mean square error as the fitness function to train the model; Mark the global optimal solution as the position of the sun; Calculate the Euclidean distances between the Sun and all celestial bodies, the gravitational force, and calculate the velocities of all celestial bodies; Update the distance between the object and the sun or the new position of the object; Calculate and compare the fitness values ​​of all celestial bodies and the sun, and update the global optimal solution until the maximum number of iterations is reached.

4. The method for designing a landing guidance law for a shipborne fixed-wing aircraft according to claim 1, characterized in that: The prediction of the motion heave value at the future key moment includes: The ship's rolling, pitching, and heaving motion data are collected every 0.05 seconds; The ship's motion posture is converted into a time series input-output pair, and the ship's posture in the next 5 seconds is predicted using the data of the past 50 time steps; The data of the three motion postures are divided into training set, test set and validation set according to the ratio of 6:2:2, and then standardized; The standardized training set is input into the KOA-LSTM hybrid forecasting model to output the predicted motion heave value.

5. The method for designing landing guidance law for a shipborne fixed-wing aircraft according to claim 1, characterized in that: The dynamic compensation correction in step 4 includes: Design compensation transfer function: ， ， ， ， in, represents the second-order compensator transfer function, represents the first-order integrator transfer function, represents the lead-lag compensator transfer function, represents the transfer function of the second-order system with zeros, represents the Laplace operator, =1, =0.8, =1.2, =1, =1.5, =1, =1, =1, =0.5, =0.1, =1, =0.3, =1, =0.5, =1, =1, =1; Set the deck motion estimation threshold to ; Calculate the compensation value of deck motion for longitudinal deviation ; in, is the predicted deck heave value in the future. It is the real-time deck heave value.

6. The method for designing landing guidance law for a shipborne fixed-wing aircraft according to claim 1, characterized in that: The calculation of the lateral and longitudinal deviations of the aircraft relative to the desired landing trajectory in step 5 includes: ignoring the deck pitch and roll conditions, and calculating the aircraft's geodetic coordinate system position. Convert to the landing coordinate system and get the coordinates ; in, The world coordinate system of the aircraft Axis position, The world coordinate system of the aircraft Axis position, The world coordinate system of the aircraft Axis position, The aircraft is in the landing coordinate system Axis position, The aircraft is in the landing coordinate system Axis position, The aircraft is in the landing coordinate system Axis position; Calculate the ideal vertical glide position value: ， in, Indicates the ideal glideslope angle of the aircraft; Calculate the height deviation without taking deck motion compensation into account: Calculate the height deviation value when considering deck motion compensation for dynamic compensation: in, The distance between the fixed-wing aircraft and the ideal landing point Axis distance.

7. The method for designing landing guidance law for a shipborne fixed-wing aircraft according to claim 1, characterized in that: Defining the relevant coordinate system in step 1 includes: World coordinate system origin Set to the sea level at the initial position of the ship, defined by the northeast, Pointing to the North Pole of the Earth, Pointing due east, The direction is consistent with the direction of gravity. Landing coordinate system origin It is an ideal landing point on the ship deck when the aircraft lands. The axis is horizontally pointing to the specified angle on the ship deck, The axis is perpendicular to the deck and faces downwards, The axis is determined according to the right-hand rule; Origin of the ship's coordinate system Fixed at the intersection of the ship's waterline and the mid-longitudinal plane, located at the bottom of the bow column. The axis points to the bow along the mid-longitudinal plane. The axis is perpendicular to the design water plane, The axis points to starboard according to the right-hand rule.

8. A landing guidance law design system for shipborne fixed-wing aircraft, characterized in that: The system comprises: Coordinate system definition unit, used to define relevant coordinate systems, including: world coordinate system, landing coordinate system, and ship body coordinate system; A height change value calculation unit is used to calculate the height change value of the ideal landing point caused by the movement of the ship deck in real time according to the position of the ideal position in the ship body coordinate system; The prediction unit is used to predict the movement and heave values ​​of the ship deck at key moments in the future using the KOA-LSTM hybrid prediction model; A compensation unit is used to design a compensation value for the longitudinal position deviation based on the predicted motion heave value and to dynamically compensate and correct the longitudinal guidance instruction; The closed-loop control unit is used to calculate the lateral and longitudinal deviations of the aircraft relative to the desired landing trajectory based on the compensated guidance instructions and the real-time position of the aircraft, thereby forming a closed-loop control.

9. A computer device, characterized in that: The invention comprises a memory and a processor, wherein a computer program is stored in the memory, and when the processor runs the computer program stored in the memory, the processor executes a method for designing a landing guidance law for a shipborne fixed-wing aircraft according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of a method for designing a landing guidance law for a shipborne fixed-wing aircraft according to any one of claims 1 to 7.

Images