A method and system for automatic landing control of a shipboard rotary-wing drone to a designated deck area

Abstract

The application discloses a kind of shipborne rotary wing unmanned aerial vehicle designated deck area automatic landing control method and system, belong to rotary wing unmanned aerial vehicle automatic control technical field.Method includes following steps: step one, based on the position of rotary wing unmanned aerial vehicle and ship deck position planning rotary wing unmanned aerial vehicle ideal landing route;Step two, based on the ideal landing route, obtain the position error of rotary wing unmanned aerial vehicle;Step three, the running posture of ship is predicted, and the posture prediction result of ship is obtained;Step four, based on the posture prediction result of ship, the predicted ideal landing route of rotary wing unmanned aerial vehicle is generated, and the position error of rotary wing unmanned aerial vehicle is based on trajectory tracking flight control of rotary wing unmanned aerial vehicle, and the automatic landing of rotary wing unmanned aerial vehicle designated deck area is completed.The application can realize the automatic landing task of shipborne rotary wing unmanned aerial vehicle in specified area on ship.

Description

Technical Field

[0001] This invention belongs to the field of automatic control technology for rotary-wing unmanned aerial vehicles (UAVs), specifically relating to an automatic landing control method and system for a shipborne rotary-wing UAV in a designated deck area. Background Technology

[0002] With the continuous advancement of my country's maritime power strategy, maritime platforms are playing an increasingly important role in military, scientific research, and rescue fields. Shipborne rotary-wing unmanned aerial vehicles (UAVs), as a highly efficient and flexible aerial platform, provide three-dimensional airspace support for ships to perform missions at sea and have broad application prospects. However, the complex maritime environment and the significant impact of wind and waves on maritime platforms make the safe landing of rotary-wing UAVs on ship platforms a pressing issue that needs to be addressed.

[0003] Research on the collaborative control of ships and UAVs is limited. For example, in August 2020, scholars Zhao Liangyu, Cheng Zhekun, Gao Fengjie, and Li Dan, in their article "Several Key Technologies for Cooperative Autonomous Landing of UAVs / Boat" published in Volume 61 (Supplement 1) of *China Shipbuilding*, pp. 156-162, introduced a guidance approach based on airborne positioning and computer vision systems, as well as basic methods for deck motion prediction and compensation. However, they did not propose detailed design methods for the attitude control of UAVs. Currently, no scholars have proposed highly reliable control algorithms for the automatic landing of shipborne rotary-wing UAVs in designated areas on ships. Summary of the Invention

[0004] This invention aims to address the shortcomings of existing technologies by proposing an automatic landing control method for a designated deck area of ​​a shipborne rotary-wing unmanned aerial vehicle (UAV), comprising the following steps:

[0005] Step 1: Plan the ideal landing route for the rotary-wing UAV based on its location and the ship's deck location;

[0006] Step 2: Based on the ideal landing path, obtain the position error of the rotary-wing UAV;

[0007] Step 3: Predict the ship's attitude and obtain the ship attitude prediction results;

[0008] Step 4: Generate the predicted ideal landing path of the rotary-wing UAV based on the ship attitude prediction results, and perform trajectory tracking flight control on the rotary-wing UAV based on the position error of the rotary-wing UAV to complete the automatic landing of the rotary-wing UAV in the designated deck area.

[0009] More preferably, the position error includes: position error during the tracking glide phase and position error during the vertical descent phase;

[0010] For a rotary-wing UAV in a world coordinate system, the coordinates are... The methods for calculating the position error during the tracking glide phase include:

[0011] ;

[0012] ;

[0013] ;

[0014] ;

[0015] In the formula, This represents the coordinates of the target point being tracked by the rotary-wing UAV in the world coordinate system. This represents the coordinates of the floating point in the world coordinate system; The starting position of the rotary-wing UAV in the entangled coordinate system Axis coordinates The expected landing rate of the drone, The time interval from the start of descent to the current time. , Representing the ideal landing path and plane angle and Plane angle; This refers to the lateral deviation. This refers to the longitudinal direction deviation; This refers to the vertical deviation.

[0016] The methods for calculating the position error during the vertical descent phase include:

[0017] When the rotorcraft UAV is within the ideal landing area, the target point it is tracking is the origin of the landing coordinate system, and the world coordinate system of the origin of the landing coordinate system is... The lateral deviation is: The longitudinal deviation is: The vertical deviation is: .

[0018] More preferably, in step three, the KOA-LSTM model is used to predict the ship's motion attitude;

[0019] Prediction methods include:

[0020] S31. Collect actual ship data and preprocess the actual ship data to obtain ship attitude data;

[0021] S32. Initialize KOA parameters: Number of planets Maximum number of iterations Tmax, hyperparameters: learning rate, upper and lower bounds of the number of neurons, control parameters , , ;in, Represents the gravitational constant; Indicates the distance attenuation coefficient; Indicates the orbital period;

[0022] S33. Initialize the parameters in step S32 with random position, orbital eccentricity and orbital period, and evaluate the fitness value of the initial population.

[0023] S34. Mark the global optimal solution as the position of the sun to obtain the current hyperparameters;

[0024] S35. Calculate the Euclidean distance between the Sun and all celestial bodies, the gravitational force between the Sun and all celestial bodies, and calculate the velocity of all celestial bodies.

[0025] S36. Update the distance between the object and the sun or the new position of the object;

[0026] S37. Calculate and compare the fitness values ​​of all celestial bodies and the Sun, and update the global optimal solution;

[0027] S38. After the iteration is completed, save the global optimal solution, which is the optimal parameter of the LSTM network;

[0028] S39. Input the ship attitude data into the LSTM network with optimal parameters to obtain the ship attitude prediction result.

[0029] More preferably, in step four, the trajectory tracking flight control method includes: using a closed-loop controller to perform trajectory tracking flight control on the rotary-wing UAV;

[0030] The input to the closed-loop controller is the desired position. and desired yaw angle The position controller obtains the desired position in the world coordinate system. Actual location and actual speed Obtain the desired tension in the body coordinate system Expected roll angle and the desired pitch angle The desired pitch angle Expected roll angle Desired yaw angle And actual Euler angles and the actual angular velocity of the machine body The input to the attitude controller yields the desired three-axis torque. The control distributor adjusts the tension according to the desired force. Desired triaxial torque To obtain the desired rotational speed The motor controller determines the desired speed. Generate the desired throttle command .

[0031] More preferably, the design method of the attitude controller includes: angle loop design and angular velocity loop design;

[0032] The methods for designing angle rings include:

[0033] ;

[0034] Expanding, we get:

[0035] ;

[0036] In the formula, This indicates the desired angular velocity in the body coordinate system. The proportional gain represents the angle error; Indicates angular error; , , They represent circumference respectively. , , Angular velocities in the roll, pitch, and yaw directions of the shaft; , , These are roll angle, pitch angle, and yaw angle, respectively. , , They represent the angles at... , , The proportional gain coefficients in the three axes;

[0037] Methods for designing angular velocity loops include:

[0038] ;

[0039] in,

[0040] ;

[0041] In the formula, The integral gain representing the angular velocity error; The differential gain representing the angular velocity error; Indicates angular velocity error; This represents the actual angular velocity input. The derivative of the angular velocity error is given.

[0042] More preferably, the design method of the position controller includes: a horizontal channel controller and a height channel controller;

[0043] The horizontal channel controller includes a position loop design and a velocity loop design;

[0044] The method for designing the position ring includes:

[0045] ;

[0046] In the formula, This indicates the desired horizontal velocity of the output. They represent , directional proportional gain coefficient;

[0047] The method for designing the velocity loop includes:

[0048] ;

[0049] get ,in , ;

[0050] Therefore, we get:

[0051] ;

[0052] in, ;

[0053] ;

[0054] In the formula, express The acceleration vector in the direction; Indicates the mass of a rotary-wing drone; This represents the lift of the UAV in equilibrium in the body coordinate system. Represents the gravitational constant; express The desired acceleration vector in the direction; The proportional gain representing the horizontal velocity error; The integral gain representing the horizontal velocity error; The differential gain representing the horizontal velocity error; Indicates horizontal velocity error; superscript -1 indicates inversion operation; superscript Represents the transpose of a matrix; Differentiating the horizontal velocity error;

[0055] The altitude channel controller includes: an altitude loop design and an altitude velocity loop design;

[0056] The method for designing the height ring includes:

[0057] ;

[0058] In the formula, This represents the desired velocity in the vertical direction of the output. The proportional gain representing the error in the vertical direction;

[0059] The method for designing the altitude velocity ring includes:

[0060] ;

[0061] Depend on replace get:

[0062] ;

[0063] ;

[0064] In the formula, express Acceleration in the axial direction; , , These represent the proportional gain, integral gain, and differential gain of the velocity error in the vertical direction, respectively. Indicates the velocity error in the vertical direction; express Differentiate; Indicates rotary-wing drone Acceleration in the axial direction.

[0065] The present invention also provides an automatic landing control system for a designated deck area of ​​a shipborne rotary-wing unmanned aerial vehicle, comprising:

[0066] The trajectory planning module is used to plan the ideal landing path for the rotary-wing UAV based on its position and the ship's deck position.

[0067] An error calculation module is used to obtain the position error of the rotary-wing UAV based on the ideal landing trajectory.

[0068] The ship attitude prediction module is used to predict the ship's operating attitude and obtain the ship attitude prediction result.

[0069] The landing control module is used to generate a predicted ideal landing path for the rotary-wing UAV based on the ship attitude prediction results, and to perform trajectory tracking flight control on the rotary-wing UAV based on the position error of the rotary-wing UAV, so as to complete the automatic landing of the rotary-wing UAV in the designated deck area.

[0070] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0071] This invention utilizes the KOA-LSTM model to predict ship motion attitude, gaining advance knowledge of the ship's dynamic changes and effectively compensating for the ship's six degrees of freedom motion. This significantly improves the landing reliability of UAVs in windy and wavy environments. Furthermore, this invention proposes a two-stage position error calculation method (descent stage and vertical landing stage), achieving more refined control by specifically correcting lateral, longitudinal, and vertical position deviations. This reduces stability issues arising from uniform control throughout the entire process, enhancing aircraft landing safety. Simultaneously, this invention provides a mechanism for determining go-around conditions, further improving landing safety and fault tolerance. Attached Figure Description

[0072] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 This is a schematic diagram of the relevant coordinate system in an embodiment of the present invention;

[0074] Figure 2 This is a diagram of the ideal landing path of a rotary-wing UAV in an embodiment of the present invention; wherein, (a) is a side view of the ideal landing path of the rotary-wing UAV; and (b) is a top view of the ideal landing path of the rotary-wing UAV.

[0075] Figure 3 This is a schematic diagram illustrating the principle of calculating the position error of a rotary-wing UAV in an embodiment of the present invention.

[0076] Figure 4 This is a flowchart of the KOA-LSTM prediction method in an embodiment of the present invention;

[0077] Figure 5 This is a diagram illustrating the roll prediction effect in an embodiment of the present invention.

[0078] Figure 6 This is a diagram illustrating the pitch prediction effect in an embodiment of the present invention.

[0079] Figure 7 This is a diagram illustrating the heave prediction effect in an embodiment of the present invention.

[0080] Figure 8 This is a schematic diagram of the closed-loop controller structure of a rotary-wing unmanned aerial vehicle in an embodiment of the present invention;

[0081] Figure 9 This is a structural diagram of the digital simulation system in an embodiment of the present invention;

[0082] Figure 10The following is a schematic diagram of the heave, pitch, and roll curves of a ship under simulated working conditions in an embodiment of the present invention; wherein, (a) is a schematic diagram of the heave curve; (b) is a schematic diagram of the pitch curve; and (c) is a schematic diagram of the roll curve.

[0083] Figure 11 This is a schematic diagram of the longitudinal deviation curve of a rotary-wing UAV under simulated working conditions in an embodiment of the present invention;

[0084] Figure 12 This is a schematic diagram of the lateral deviation curve of a rotary-wing UAV under simulated working conditions in an embodiment of the present invention;

[0085] Figure 13 This is a schematic diagram of the yaw angle curve of a rotary-wing UAV under simulated working conditions in an embodiment of the present invention;

[0086] Figure 14 This is a schematic diagram of the pitch angle curve of a rotary-wing UAV under simulated working conditions in an embodiment of the present invention;

[0087] Figure 15 This is a schematic diagram of the roll angle curve of a rotary-wing UAV under simulated working conditions in an embodiment of the present invention;

[0088] Figure 16 This is a schematic diagram of the heave, pitch, and roll curves of a ship under simulation condition two in this embodiment of the invention; wherein, (a) is a schematic diagram of the heave curve; (b) is a schematic diagram of the pitch curve; and (c) is a schematic diagram of the roll curve.

[0089] Figure 17 This is a schematic diagram of the longitudinal deviation curve of a rotary-wing UAV under simulated working condition two in an embodiment of the present invention;

[0090] Figure 18 This is a schematic diagram of the lateral deviation curve of a rotary-wing UAV under simulated working condition two in an embodiment of the present invention;

[0091] Figure 19 This is a schematic diagram of the yaw angle curve of a rotary-wing UAV under simulated working condition two in an embodiment of the present invention;

[0092] Figure 20 This is a schematic diagram of the pitch angle curve of a rotary-wing UAV under simulated working condition two in an embodiment of the present invention;

[0093] Figure 21 This is a schematic diagram of the roll angle curve of a rotary-wing UAV under simulation condition two in an embodiment of the present invention. Detailed Implementation

[0094] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0095] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0096] First, the relevant coordinate systems will be explained, as follows: Figure 1 As shown, it includes: world coordinate system, ship body coordinate system, landing coordinate system, and associated world coordinate system.

[0097] (1) World coordinate system :

[0098] The origin of the world coordinate system is located at a certain point on the Earth. The positive direction of the axis points north. The positive direction of the axis points eastward. The positive direction of the axis is perpendicular to the sea level and downwards.

[0099] (2) Ship body coordinate system :

[0100] The origin of the ship's coordinate system is located at the ship's sway center. The positive direction of the axis points towards the bow. The axis is pointing in the starboard direction of the ship. The positive axis is perpendicular to the deck and points towards the seabed.

[0101] (3) Landing coordinate system :

[0102] The origin of the landing coordinate system is located at the geometric center of the rectangular landing area. The positive direction of the axis points towards the bow. The axis is pointing in the starboard direction of the ship. The positive axis is perpendicular to the deck and points towards the seabed.

[0103] (4) Involving the world coordinate system :

[0104] origin The point a certain distance above the origin of the landing coordinate system is the leveling point. This distance is shown in [reference needed]. Figure 2 In the ideal landing path of an aircraft, the directions of the three coordinate axes (d) are consistent with the directions of the three coordinate axes of the world coordinate system.

[0105] (5) Body coordinate system :

[0106] Origin of the coordinate system of the rotorcraft UAV Located at the center of mass of the UAV. Coordinate system. The positive direction of the axis points towards the nose of the drone, that is, directly in front of the drone; The positive direction of the axis points to the right side of the drone; The positive axis points vertically downwards, towards the ground.

[0107] Example 1:

[0108] Step 1: Plan the ideal landing route for the rotary-wing drone based on the rotary-wing drone and the ship deck.

[0109] In this embodiment, the landing area for the rotary-wing UAV is the upper surface of a cuboid on the ship's deck. The ideal landing point for the shipborne rotary-wing UAV during descent is the geometric center of the cuboid's upper surface. It descends at a preset glide angle, and after reaching a certain height above the landing area, it vertically lands within the designated landing area to complete the recovery mission. The ideal landing path for the rotary-wing UAV in this embodiment is as follows: Figure 2 (a) and Figure 2 As shown in (b).

[0110] like Figure 2 As shown in (b), during descent, the rotary-wing UAV approaches the ideal landing point from the left rear of the ship, and descends vertically until it stops above the ideal landing point. Figure 2 In (a), this embodiment defines the ideal landing path and the landing coordinate system. Angle between axes =30 degrees, at Figure 2 In (b), this embodiment defines the angle between the ideal landing path and the horizontal plane. =45 degrees, distance above the ideal landing point =5 meters. Additionally, in this embodiment, the rotary-wing UAV is designed to fly to approximately 50 meters diagonally behind the stern of the ship to begin preparing for landing.

[0111] Step 2: Calculate the position error of the rotary-wing UAV based on the ideal landing path.

[0112] This embodiment divides the landing process of the rotary-wing UAV into two stages based on the ideal landing trajectory: the tracking descent stage and the vertical landing stage. The position error of the rotary-wing UAV is calculated in each stage, and the calculation method is as follows:

[0113] Step 1: Method for calculating position error during the descent phase.

[0114] During the descent phase of a rotary-wing UAV's tracking, in order for the UAV to accurately track the ideal glide path, it is necessary to acquire the tracking points on the ideal glide path in real time. The principle for calculating the position error during the UAV's descent in this embodiment is as follows: Figure 3 As shown, the ideal landing path in this embodiment is represented as follows: Its coordinates in the landing coordinate system are the same as The included angle between the planes is ,and The included angle between the planes is Let the initial position of the rotorcraft UAV when it begins to land be... The point, whose coordinates in the world coordinate system are... The target point to be tracked must always remain on the designed ideal glide path and at a constant rate. Decline. That is... Figure 3 Point A in the middle. Let the coordinates of the floating point in the world coordinate system be... Then the coordinates of point A in the world system are:

[0115] ; (1)

[0116] In the formula, This represents the coordinates of the target point being tracked by the rotary-wing UAV in the world coordinate system. The expected landing rate of the drone; This is the time interval from the start of descent to the current time. This refers to the lateral deviation. This refers to the longitudinal direction deviation; This represents the vertical deviation.

[0117] Let the coordinates of the rotary-wing UAV in the world coordinate system be... Then the lateral deviation is: The longitudinal deviation is: The vertical deviation is: .

[0118] Step 2: Method for calculating position error during vertical descent.

[0119] Above the ideal landing point of the rotary-wing drone At the 5-meter mark, the following situations apply, and corresponding actions are taken accordingly: If the rotorcraft's lateral and longitudinal positions are not within the ideal landing area, the rotorcraft will attempt a go-around. If the rotorcraft's lateral and longitudinal positions are within the ideal landing area, the rotorcraft will descend vertically into the recovery capsule, completing the landing operation.

[0120] If the rotary-wing UAV is within the ideal landing area, then its tracking target point is the origin of the landing coordinate system, and the world coordinate system of the origin of the landing coordinate system is... Then the lateral deviation is: The longitudinal deviation is: The vertical deviation is: .

[0121] Step 3: Predict the ship's motion attitude and obtain the ship attitude prediction results.

[0122] Specifically, the KOA-LSTM model is used to predict the ship's motion attitude. First, actual ship data is collected, with data on the ship's roll, pitch, and heave motion attitudes collected every 0.05 seconds, for a total of 20,000 data points, or 1000 seconds of data. The collected data is then processed: the ship's motion attitude is transformed into input-output data pairs with clear temporal correlation, forming a supervised learning format suitable for training the LSTM neural network. In this embodiment, the task is to predict the ship's attitude 5 seconds into the future using data from the past 50 time steps; therefore, each sample needs to include attitude data from the past 50 time steps and the future 100 time steps. The attitude data for the three motions are divided into training, testing, and validation sets in a 6:2:2 ratio, and standardized to eliminate the influence of dimensions. Finally, the data is reconstructed into the three-dimensional structure required by the LSTM network, including the number of samples, time steps, and number of features, to obtain the ship's attitude data.

[0123] When performing time series prediction of ship pose, the selection of hyperparameters plays a decisive role in the prediction accuracy of the LSTM network. This embodiment combines the Kepler Optimization Algorithm (KOA) with LSTM to construct a KOA-LSTM model. The structure of the KOA-LSTM model is as follows: Figure 4 As shown, the specific process is as follows:

[0124] S31. Prepare ship attitude data according to the above method, train and validate the KOA-LSTM model using the training set and validation set, and perform prediction tests on the KOA-LSTM model using the test set.

[0125] S32. Initialize KOA parameters: Number of planets Maximum number of iterations Tmax, upper and lower bounds of hyperparameters (learning rate, number of neurons), and control parameters. , , .in, This represents the gravitational constant and controls the update step size and convergence speed. This represents the distance decay coefficient, used to adjust the balance between exploration and convergence; It represents the orbital period, used to determine the flexibility and robustness of the search trajectory.

[0126] S33. Initialize the celestial population with random position, orbital eccentricity and orbital period (i.e. the parameters in step S32), and evaluate the fitness value of the initial population, using the root mean square error (RMSE) as the fitness function.

[0127] S34. Mark the global optimal solution as the position of the sun, which is the current hyperparameter selection.

[0128] S35. Calculate the Euclidean distance between the Sun and all celestial bodies, the gravitational force between them, and the velocity of all celestial bodies.

[0129] S36. Update the distance between the object and the sun or the new position of the object.

[0130] S37. Calculate and compare the fitness values ​​of all celestial bodies and the Sun, and update the global optimal solution.

[0131] S38. After the iteration is completed, save the global optimal solution, which is the optimal parameter selection for the LSTM network.

[0132] S39. Input the ship attitude data into the LSTM network with optimal parameters to obtain the ship attitude prediction result.

[0133] like Figure 5 , Figure 6 , Figure 7 As shown, the prediction results for the test set data of ship roll, pitch, and heave motions are presented respectively. Among them, the RMSE and MAE for roll prediction are 8.29% and 6.28%, respectively; the RMSE and MAE for pitch prediction are 2.85% and 2.31%, respectively; and the RMSE and MAE for heave prediction are 2.19% and 1.73%, respectively.

[0134] Step 4: Generate the predicted ideal landing path of the rotorcraft UAV based on the predicted state of the ship, and perform trajectory tracking flight control on the rotorcraft UAV based on the position error of the rotorcraft UAV to complete the automatic landing of the rotorcraft UAV in the designated deck area.

[0135] In this embodiment, the flight control algorithm of the rotary-wing UAV is mainly divided into two parts: position control and attitude control. The closed-loop controller structure of the rotary-wing UAV is as follows: Figure 8 As shown. In the overall block diagram of the closed-loop controller, the given input is the desired position. and desired yaw angle The position controller obtains the desired position in the world coordinate system. Actual location and actual speed Find the desired tension in the body coordinate system. Expected roll angle and the desired pitch angle ,Will , , Actual Euler angles and the actual angular velocity of the machine body The input is fed into the attitude controller to calculate the desired triaxial torque. The control distributor adjusts the tension according to the desired force. Desired triaxial torque Find the desired rotational speed The motor controller determines the desired speed. Generate the desired throttle command .

[0136] To design a closed-loop controller, it is necessary to linearize it near the equilibrium point. When a rotary-wing UAV is hovering, since all states remain unchanged, it is at an equilibrium point. At this point, the Euler angles (except for...) Since all are close to 0, therefore: .in, Indicates the mass of a rotary-wing drone; Represents the gravitational constant; This represents the transformation matrix between the body's angular velocity and Euler angles; Represents a 3D identity matrix; , , These are roll angle, pitch angle, and yaw angle, respectively. This represents the lift of the UAV in equilibrium in the body coordinate system. Therefore, a rotorcraft UAV model can be established as follows:

[0137] ; (2)

[0138] ; (3)

[0139] ; (4)

[0140] ; (5)

[0141] In the formula, express Differentiate; This represents the speed of the drone in the world coordinate system. This represents the acceleration of the drone in the world coordinate system. Represents the Euler angular rate; This represents the actual angular velocity input. This represents the total torque provided by the rotor in the lower coordinate system; The inertia matrix is ​​represented in the body coordinate system. This represents the angular acceleration of the rotorcraft UAV in the body coordinate system.

[0142] The attitude controller design method includes: In this embodiment, the attitude controller adopts cascade PID control. The outermost PID controls the angle, which is generally controlled by P. The innermost PID controls the angular velocity.

[0143] (1) Angle loop: The input is the angle error of three Euler angles, and the output is the expected angular velocity in the body coordinate system. Table 1 shows the symbol explanation of the angle loop.

[0144] Table 1 Explanation of Angle Ring Symbols

[0145]

[0146] In this embodiment, the angle loop uses proportional control, as shown in the following formula:

[0147] (6)

[0148] Expanding formula (6) into matrix form gives:

[0149] ; (7)

[0150] In the formula, , , They represent circumference respectively. , , Angular velocities in the roll, pitch, and yaw directions of the shaft; , , They represent the angles at... , , The proportional gain coefficients in the three axes reflect the angular error response capability in their respective directions.

[0151] (2) Angular velocity loop: The input is the desired angular velocity, and the output is the desired triaxial torque. Table 2 shows the symbol explanation of the angular velocity loop.

[0152] Table 2 Explanation of Angular Velocity Ring Symbols

[0153]

[0154] The angular velocity loop uses PID control, with the following formula:

[0155] ; (8)

[0156] In the formula, The derivative of the angular velocity error is given.

[0157] The design method of the position controller includes: the position controller is designed with cascade PID control, the outer loop controls the position using only P control, and the inner loop controls the speed using PID control.

[0158] Based on the model of the rotary-wing UAV (i.e., formula (2) - formula (5)), it can be seen that and The dynamic equations for the direction are related to the three Euler angles, and Since the axis does not include Euler angles, they are designed separately. and Direction is controlled by a single controller. A separate controller is designed for each direction. and The direction controller is also known as the horizontal channel controller. The direction controller is also known as the height channel controller.

[0159] (1) Horizontal channel controller: The input is the desired input in the world coordinate system. , The direction of the position outputs the desired roll angle. and the desired pitch angle .

[0160] ① Position ring, Table 3 shows the symbol for the horizontal position ring.

[0161] The position loop uses proportional control, with the following formula:

[0162] ; (9)

[0163] In the formula, They represent , directional proportional gain coefficient; The desired horizontal speed for the output.

[0164] ② Velocity loop, Table 3 shows the symbols for the horizontal velocity loop.

[0165] Table 3 Explanation of Horizontal Velocity Ring Symbols

[0166]

[0167] For horizontal velocity, the following relationship holds:

[0168] (10)

[0169] make , have:

[0170] (11)

[0171] so,

[0172] (12)

[0173] In the formula, express The acceleration vector in the direction; express The desired acceleration vector in the direction; the superscript -1 indicates the inverse operation.

[0174] The speed loop uses PID control, with the following formula:

[0175] ; (13)

[0176] In the formula, The derivative of the horizontal velocity error is given.

[0177] Then, according to formula (12), we can obtain That is, to obtain the desired roll angle. and the desired pitch angle .

[0178] (2) Height Channel Controller: The height channel input includes the desired vertical height, the actual vertical height, and the actual vertical direction. The axial velocity outputs the desired thrust of the drone.

[0179] ① Height ring design: Table 4 shows the height ring symbols.

[0180] Table 4 Explanation of Height Ring Symbols

[0181]

[0182] Similarly, the vertical position loop also uses proportional control, and the output of this position loop is the desired velocity in the vertical direction. :

[0183] (14)

[0184] ② Vertical velocity ring design, Table 5 shows the symbols for the vertical velocity ring.

[0185] Table 5 Explanation of Vertical Velocity Loop Symbols

[0186]

[0187] The input is the desired vertical velocity, and the output is the desired pulling force.

[0188] ; (15)

[0189] In the formula, Indicates rotary-wing drone Acceleration in the axial direction.

[0190] When designing the controller, Replace with Therefore, the following equation holds true:

[0191] ; (16)

[0192] ; (17)

[0193] In the formula, express Differentiate it.

[0194] The position controller and attitude controller were designed using the methods described above.

[0195] The control distributor controls the desired force output by the position controller. and the desired torque output by the attitude controller Adjust the rotation speed of each rotor of the drone. .

[0196] The relationship between the drone's rotational speed and the force and torque is as follows:

[0197] ; (18)

[0198] In the formula, , , , For different rotor speeds, , , They are respectively exist , , The component of the torque in the positive direction of the axis, Let be the lift coefficient of the rotor. The vertical distance from the rotor centerline to the center of mass of the UAV is denoted as . is the rotor torque coefficient.

[0199] The motor controller generates motor throttle commands based on the rotor speed output from the control distributor. The motor model is as follows:

[0200] ; (19)

[0201] ; (20)

[0202] In the formula, This is the steady-state speed of the motor. For speed parameters, is the rotational speed constant.

[0203] You can get throttle commands With the steady-state speed of the motor The relationship is:

[0204] ; (twenty one)

[0205] Therefore, we can obtain:

[0206] . (twenty two)

[0207] Example 2:

[0208] This embodiment designs an automatic landing simulation system for a shipborne rotary-wing unmanned aerial vehicle (UAV). The structure of the digital simulation system is as follows: Figure 9 As shown, the digital simulation system includes: an aircraft motion module (UAV dynamics model, UAV kinematics model), a ship motion module, a prediction module, a sea state setting module, an initial parameter setting module, a guidance law module (which inputs the UAV's ideal position and the UAV's actual position obtained by radar into the guidance law computer and outputs guidance commands), and a flight control module (autopilot instrument), etc.

[0209] The ship motion module includes modules for the hull, waterjet propulsion, waves, wind, current, viscous forces, and inertial forces. These modules drive the ship's motion by applying forces and torques to the hull, thereby outputting the ship's position, attitude, and 6-DOF state. Based on this, the prediction and coordinate transformation modules generate the ship's ideal landing point and ideal landing path in the world coordinate system using the state information from the ship motion module. The guidance law module transforms the ideal landing point and ideal landing path into lateral, longitudinal, and vertical positional deviations provided by the rotorcraft UAV. The flight control algorithm calculates the aircraft's control variables based on the state deviations, thereby eliminating the tracking deviations of the rotorcraft UAV.

[0210] Simulation Condition 1: Sea state 4, ship speed 6 knots, sailing against the waves, using irregular waves for wave simulation, the ship motion simulation curve is as follows. Figure 10 (a) Figure 10 (b) Figure 10As shown in (c), the simulation diagrams (heave, pitch, roll) demonstrate the typical 6-DOF motion characteristics of a ship under sea state 4, influenced by waves and wind. Under this sea state, the ship exhibits large-amplitude periodic oscillations (the pitch angle can reach up to 8°). The position and attitude deviation curves during the landing process of the rotary-wing UAV are shown in Figure [image missing]. Figures 11-15 As shown, the lateral and longitudinal deviations represent the dynamic position deviation curves of the UAV when tracking the target point in the ideal landing area of ​​the ship. Although the ship's motion is violent, the deviation remains within a controllable range, demonstrating that the proposed flight control algorithm possesses excellent real-time response and dynamic compensation capabilities. Simulation results show that the UAV can generate a dynamically adaptive desired trajectory based on the predicted ship state and track it in real time, effectively coping with the ship's irregular motion. Especially during landing, the UAV can quickly adjust its own motion state, ensuring precise landing operations under the uncertainty of the ship's deck motion, demonstrating high adaptability and robustness.

[0211] Simulation Condition 2: Sea state 4, ship speed 10 knots, sailing against the waves, using irregular waves for wave simulation, the ship motion simulation curve is as follows. Figure 16 (a) Figure 16 (b) Figure 16 As shown in (c), compared to the simulation at 6 knots, the frequency and amplitude of ship motion are significantly increased under this sea state, simulating more stringent landing conditions. The position and attitude deviation curves during the landing process of the rotary-wing UAV are shown in Figure [image missing]. Figures 17-21 As shown, although the violent movement of the ship's deck brings greater dynamic disturbances, the UAV deviation curve remains stable and the fluctuations are controlled, fully demonstrating the strong adaptability and real-time compensation capability of the flight control algorithm. The attitude deviation curve shows that the UAV's attitude response is rapid and the convergence is good. Although the amplitude of the disturbance generated by the ship increases at higher speeds, the UAV's attitude can always respond quickly and tend to stabilize, indicating that the control system has excellent robustness and fast convergence. Overall, the simulation results prove that the control algorithm proposed in this invention, based on ship state prediction and flight error coupling compensation, can achieve high-precision tracking control of UAV trajectory and attitude in higher speeds and more complex sea states, and has significant engineering adaptability and safety advantages.

[0212] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for automatic landing control of a shipborne rotary-wing unmanned aerial vehicle in a designated deck area, characterized in that, Includes the following steps: Step 1: Plan the ideal landing route for the rotary-wing UAV based on its location and the ship's deck location; Step 2: Based on the ideal landing path, obtain the position error of the rotary-wing UAV; Step 3: Predict the ship's attitude and obtain the ship attitude prediction results; Step 4: Generate the predicted ideal landing path of the rotary-wing UAV based on the ship attitude prediction results, and perform trajectory tracking flight control on the rotary-wing UAV based on the position error of the rotary-wing UAV to complete the automatic landing of the rotary-wing UAV in the designated deck area. In step three, the KOA-LSTM model is used to predict the ship's motion attitude; Prediction methods include: S31. Collect actual ship data and preprocess the actual ship data to obtain ship attitude data; S32. Initialize KOA parameters: Number of planets Maximum number of iterations Tmax, hyperparameters: learning rate, upper and lower bounds of the number of neurons, control parameters , , ;in, Represents the gravitational constant; Indicates the distance attenuation coefficient; Indicates the orbital period; S33. Initialize the parameters in step S32 with random position, orbital eccentricity and orbital period, and evaluate the fitness value of the initial population. S34. Mark the global optimal solution as the position of the sun to obtain the current hyperparameters; S35. Calculate the Euclidean distance between the Sun and all celestial bodies, the gravitational force between the Sun and all celestial bodies, and calculate the velocity of all celestial bodies. S36. Update the distance between the object and the sun or the new position of the object; S37. Calculate and compare the fitness values ​​of all celestial bodies and the Sun, and update the global optimal solution; S38. After the iteration is completed, save the global optimal solution, which is the optimal parameter of the LSTM network; S39. Input the ship attitude data into an LSTM network with optimal parameters to obtain the ship attitude prediction result; In step four, the trajectory tracking flight control method includes: using a closed-loop controller to perform trajectory tracking flight control on the rotary-wing UAV; The input to the closed-loop controller is the desired position. and desired yaw angle The position controller obtains the desired position in the world coordinate system. Actual location and actual speed Obtain the desired tension in the body coordinate system Expected roll angle and the desired pitch angle The desired pitch angle Expected roll angle Desired yaw angle And actual Euler angles and the actual angular velocity of the machine body The input to the attitude controller yields the desired three-axis torque. The control distributor adjusts the tension according to the desired force. Desired triaxial torque To obtain the desired rotational speed The motor controller determines the desired speed. Generate the desired throttle command .

2. The automatic landing control method for a designated deck area of ​​a shipborne rotary-wing unmanned aerial vehicle according to claim 1, characterized in that, The position error includes: position error during the tracking glide phase and position error during the vertical descent phase; For a rotary-wing UAV in a world coordinate system, the coordinates are... The methods for calculating the position error during the tracking glide phase include: ； ； ； ； In the formula, This represents the coordinates of the target point being tracked by the rotary-wing UAV in the world coordinate system. This represents the coordinates of the floating point in the world coordinate system; The starting position of the rotary-wing UAV in the entangled coordinate system Axis coordinates The expected landing rate of the drone, The time interval from the start of descent to the current time. , Representing the ideal landing path and plane angle and Plane angle; This refers to the lateral deviation. This refers to the longitudinal direction deviation; This refers to the vertical deviation. The methods for calculating the position error during the vertical descent phase include: When the rotary-wing UAV is within the ideal landing area, the target point being tracked by the UAV is the origin of the landing coordinate system, and the world coordinate system of the origin of the landing coordinate system is... The lateral deviation is: The longitudinal deviation is: The vertical deviation is: .

3. The automatic landing control method for a designated deck area of ​​a shipborne rotary-wing unmanned aerial vehicle according to claim 1, characterized in that, The design method of the attitude controller includes: angle loop design and angular velocity loop design; The methods for designing angle rings include: ； Expanding, we get: ； In the formula, This indicates the desired angular velocity in the body coordinate system. The proportional gain represents the angle error; Indicates angular error; , , They represent circumference respectively. , , Angular velocities in the roll, pitch, and yaw directions of the shaft; , , These are roll angle, pitch angle, and yaw angle, respectively. , , They represent the angles at... , , The proportional gain coefficients in the three axes; Methods for designing angular velocity loops include: ； in, ； In the formula, The integral gain representing the angular velocity error; The differential gain representing the angular velocity error; Indicates angular velocity error; This represents the actual angular velocity input. The derivative of the angular velocity error is given.

4. The automatic landing control method for a designated deck area of ​​a shipborne rotary-wing unmanned aerial vehicle according to claim 3, characterized in that, The design method for the position controller includes: a horizontal channel controller and a height channel controller; The horizontal channel controller includes a position loop design and a velocity loop design; The method for designing the position ring includes: ； In the formula, This indicates the desired horizontal velocity of the output. They represent , directional proportional gain coefficient; The method for designing the velocity loop includes: ； get ,in , ; Therefore, we get: ； in, ; ； In the formula, express The acceleration vector in the direction; Indicates the mass of a rotary-wing drone; This represents the lift of the UAV in equilibrium in the body coordinate system. Represents the gravitational constant; express The desired acceleration vector in the direction; The proportional gain representing the horizontal velocity error; The integral gain representing the horizontal velocity error; The differential gain representing the horizontal velocity error; Indicates horizontal velocity error; superscript -1 indicates inversion operation; superscript Represents the transpose of a matrix; Differentiating the horizontal velocity error; The altitude channel controller includes: an altitude loop design and an altitude velocity loop design; The method for designing the height ring includes: ； In the formula, This represents the desired velocity in the vertical direction of the output. The proportional gain representing the error in the vertical direction; The method for designing the altitude velocity ring includes: ； Depend on replace get: ； ； In the formula, express Acceleration in the axial direction; , , These represent the proportional gain, integral gain, and differential gain of the velocity error in the vertical direction, respectively. Indicates the velocity error in the vertical direction; express Differentiate; Indicates rotary-wing drone Acceleration in the axial direction.

5. An automatic landing control system for a designated deck area of ​​a shipborne rotary-wing unmanned aerial vehicle, the control system being used to implement the method as described in any one of claims 1-4, characterized in that, include: The trajectory planning module is used to plan the ideal landing path for the rotary-wing UAV based on its position and the ship's deck position. An error calculation module is used to obtain the position error of the rotary-wing UAV based on the ideal landing trajectory. The ship attitude prediction module is used to predict the ship's operating attitude and obtain the ship attitude prediction result. The landing control module is used to generate a predicted ideal landing path for the rotary-wing UAV based on the ship attitude prediction results, and to perform trajectory tracking flight control on the rotary-wing UAV based on the position error of the rotary-wing UAV, so as to complete the automatic landing of the rotary-wing UAV in the designated deck area.

Images