A roll angle stabilization control method for unmanned aerial vehicles

Abstract

The application discloses a kind of for unmanned aerial vehicle roll angle's stabilizing control method, belong to the field of quadrotor aircraft attitude angle stabilizing control, comprising: constructing three-order linearized dynamics model;Three-order linearized dynamics model is handled using pole assignment technique, to obtain stable state matrix;Jordan canonical form is constructed based on the eigenvalue component of stable state matrix, to obtain positive eigenvalue matrix and real matrix, and deduce high dynamic output matrix;High dynamic roll angle stabilizing control signal is designed based on positive eigenvalue matrix and high dynamic output matrix;High dynamic roll angle stabilizing control signal is used to solve the command roll torque;The command roll torque is input coupling roll torque output characteristic model, to obtain unmanned aerial vehicle roll channel actual roll torque value.The application improves the high dynamic convergence ability of roll angle, so that unmanned aerial vehicle can be faster, more stable to realize the stabilizing control of roll angle under complex environment.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of attitude angle stabilization control of quadrotor aircraft, and particularly relates to a stabilization control method for roll angle of unmanned aerial vehicle. BACKGROUND

[0002] The wide application of unmanned aerial vehicles in complex dynamic scenes such as logistics inspection, disaster rescue and aerial photography and surveying puts increasingly stringent requirements on the high-maneuvering flight capability and rapid attitude stabilization capability of the unmanned aerial vehicles. The dynamic response speed and stabilization accuracy of roll angle control, as a key link of attitude control of unmanned aerial vehicles, directly determine the safety and task efficiency of the vehicles in high dynamic tasks such as high-speed evasion and emergency steering. However, the existing roll angle control methods face significant challenges: on the one hand, the traditional design based on a simplified model (such as a second-order system) is difficult to accurately represent and compensate the complex coupling dynamics of the driving system in the roll channel, resulting in response lag, overshoot and even instability under high dynamic maneuvering; on the other hand, the existing stabilization controllers often rely on tedious parameter tuning to balance the response speed and robustness, and are prone to introduce static errors, making it difficult to achieve high dynamic convergence while realizing static error suppression. Therefore, it is urgent to develop a new roll angle stabilization control method that can deeply analyze the roll coupling dynamics and effectively integrate the high dynamic convergence and static error suppression capability to meet the stringent requirements of modern unmanned aerial vehicles on flight quality under extreme working conditions. SUMMARY

[0003] To solve the above technical problems, the application provides a stabilization control method for roll angle of unmanned aerial vehicle to solve the problems existing in the prior art.

[0004] To achieve the above purpose, the application provides a stabilization control method for roll angle of unmanned aerial vehicle, comprising:

[0005] constructing a third-order linearized dynamics model based on the roll channel of the unmanned aerial vehicle;

[0006] processing an unstable state matrix in the third-order linearized dynamics model by using pole placement technology to obtain a stable state matrix, and constructing a state stable dynamics model of the roll channel;

[0007] constructing a Jordan standard form based on the eigenvalue components of the stable state matrix to obtain a positive eigenvalue matrix and a real number matrix;

[0008] based on the positive eigenvalue matrix, the real number matrix and an auxiliary matrix, deriving a high dynamic output matrix through equation transformation;

[0009] designing a high dynamic roll angle stabilization control signal based on the positive eigenvalue matrix and the high dynamic output matrix;

[0010] The command roll torque is derived by inversely solving the high dynamic roll angle stabilization control signal.

[0011] By coupling the command roll torque input with the roll torque output characteristic model, the actual roll torque value of the unmanned aerial vehicle roll channel at each sampling time is obtained.

[0012] Optionally, the process of constructing a third-order linearized dynamic model based on the roll channel of the unmanned aerial vehicle includes:

[0013] A coupled rolling torque output characteristic model is established based on the coupled rolling torque and command rolling torque of the motor drive system.

[0014] Establish a linearized relationship between roll angle, roll angular velocity, and roll torque;

[0015] By defining the roll channel state vector, the unstable state matrix, and the constant control vector, the coupled roll torque output characteristic model is combined with the linearized relationship to obtain a third-order linearized dynamic model based on the coupled roll torque.

[0016] Optionally, the expression for the coupled rolling torque output characteristic model is:

[0017] ;

[0018] In the formula, for The derivative of For the bandwidth of the motor drive system, The moment of inertia of the rolling channel. for The reciprocal, The coupled rolling torque generated in the rolling channel by the motor drive system of the unmanned aerial vehicle. This is the command rolling torque.

[0019] Optionally, the expression for the third-order linearized dynamic model based on the coupled rolling moment is:

[0020] ;

[0021] In the formula, The matrix represents the unstable state. For constant control vectors, This is the roll channel state vector. for The derivative of for The reciprocal, This is the command rolling torque.

[0022] Optionally, the process of constructing a Jordan canonical form based on the eigenvalue components of the steady-state matrix, to obtain a positive eigenvalue matrix and a real matrix, comprises:

[0023] Classifying the eigenvalues of the steady-state matrix to obtain negative real roots and conjugate complex roots;

[0024] Constructing a positive eigenvalue matrix from the reciprocals of the negative real roots;

[0025] Constructing a complex Jordan block from the reciprocals of the conjugate complex roots;

[0026] Constructing a Jordan canonical form based on the positive eigenvalue matrix and the complex Jordan block;

[0027] Obtaining the real matrix by solving the conversion relationship between the real matrix and the Jordan canonical form.

[0028] Optionally, the process of deriving a high-dynamic output matrix based on the positive eigenvalue matrix, the real matrix, and an auxiliary matrix through equation transformation comprises:

[0029] Transposing the equation of the real matrix and introducing an auxiliary matrix, and eliminating intermediate variables through matrix operations to obtain a simplified equation containing a positive eigenvalue matrix;

[0030] Defining a high-dynamic output matrix based on the simplified equation.

[0031] Optionally, the method further comprises constructing a minimum-phase dynamic system for controller design based on the high-dynamic output matrix and the state-stable dynamics model of the roll channel;

[0032] wherein the expression of the minimum-phase dynamic system for controller design is:

[0033] ;

[0034] wherein, is the derivative of , is the output variable, is a positive eigenvalue matrix, T is a transpose, is a high-dynamic output matrix, is a constant control vector, is a high-dynamic roll angle stabilization control signal that needs to be further designed.

[0035] Optionally, the expression of the designed high-dynamic roll angle stabilization control signal is:

[0036] ;

[0037] wherein, For the high dynamic roll angle stabilization control signal that requires further design, For proportional sedation items, Here, T is the integral stabilization term, and T is the transpose. For high dynamic range output matrix, For a certain moment, For design parameters that are greater than zero, This is the roll channel state vector. It is a positive eigenvalue matrix. Indicates from the initial time to The points, This is a constant control vector.

[0038] The present invention also provides a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements a stabilization control method for the roll angle of an unmanned aerial vehicle.

[0039] The present invention also provides a storage medium storing a computer program that, when executed by a processor, implements a stabilization control method for the roll angle of an unmanned aerial vehicle.

[0040] Compared with the prior art, the present invention has the following advantages and technical effects:

[0041] First, by constructing a third-order linearized dynamic model that includes coupled roll torque, the complex dynamic characteristics of the roll channel are more accurately characterized. Second, the pole placement technique is used to handle the unstable state matrix, effectively improving the stability of the system. Third, by constructing the Jordan canonical form and deriving the high dynamic output matrix, the placement of conjugate complex roots is realized, significantly improving the dynamic response speed of the roll angle control. Finally, a composite control signal containing proportional and integral terms is introduced, which effectively suppresses static error while ensuring fast convergence. Attached Figure Description

[0042] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0043] Figure 1 This is a flowchart illustrating the overall process of the high dynamic stabilization control method for the roll angle of unmanned aerial vehicles according to an embodiment of the present invention.

[0044] Figure 2 This is a comparison chart of the roll angle convergence characteristics of embodiments of the present invention. Detailed Implementation

[0045] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0046] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0047] like Figure 1 As shown, this embodiment provides a stabilization control method for the roll angle of an unmanned aerial vehicle, including the following steps:

[0048] Step S1: Construct a third-order linearized dynamic model based on coupled rolling torque for the roll channel of the unmanned aerial vehicle.

[0049] Furthermore, the process of constructing a third-order linearized dynamic model includes:

[0050] First, the coupled rolling torque generated by the motor drive system on the unmanned aerial vehicle in the rolling channel. and command rolling torque The following dynamic equations (coupled roll torque output characteristic model) are satisfied between them:

[0051] (1)

[0052] in, for The derivative of For the bandwidth of the motor drive system, The moment of inertia of the rolling channel. for The reciprocal of.

[0053] Next, the linearized dynamic model of the unmanned aerial vehicle's roll channel dynamics near the equilibrium point can be expressed as:

[0054] (2)

[0055] in, For roll angle, for The derivative of For the roll angular velocity, for The derivative of This represents the actual rolling torque. To embed the rolling torque output characteristic model into the linearized dynamics model, it is necessary to define the rolling channel state vector. Unstable state matrix and the constant control vector , the specific expression is as formula (3).

[0056] (3)

[0057] Therefore, under the condition that is established, a third-order linearized dynamics model based on the coupling roll moment can be obtained:

[0058] (4)

[0059] Step S2, the unstable state matrix existing in the third-order linearized dynamics model based on the coupling roll moment is processed by using the pole placement technique, so as to construct a state stable dynamics model facing the roll channel.

[0060] Further, the process of constructing the state stable dynamics model facing the roll channel includes:

[0061] The obvious mark of the unstable state matrix is that its eigenvalue contains zero or positive number. If the controller directly controls the motion trajectory of the roll angle in the presence of the unstable state matrix , it is difficult to achieve the predetermined convergence characteristics, and even there is a risk of divergence of the roll angle. Therefore, it is necessary to further process the unstable state matrix by using the pole placement technique, so that it becomes a stable state matrix . The key operation steps of the pole placement technique are as follows:

[0062] (1) select appropriate eigenvalues (negative roots or conjugate negative roots) for the unstable state matrix ;

[0063] (2) solve the gain matrix by using the pole placement function provided by MATLAB;

[0064] (3) solve the stable state matrix by using the equation .

[0065] Next, only need to introduce the data item in the third-order linearized dynamics model based on the coupling roll moment, so as to obtain the state stable dynamics model facing the roll channel:

[0066] (5)

[0067] Step S3, construct the Jordan standard type according to the eigenvalue component of the stable state matrix, so as to obtain the positive eigenvalue matrix and the real number matrix. The specific implementation process includes:

[0068] This invention provides a general method for solving external eigenvalue matrices and real matrices, which is unaffected by the steady-state matrix. Dimensionality constraint. Next, we derive the state matrix of dimension n. The corresponding methods for solving the external eigenvalue matrix and the real matrix.

[0069] To facilitate the construction of the Jordan canonical form, it is first necessary to analyze the stable state matrix. (symbol The eigenvalues ​​of a matrix of dimension n are classified into two main categories: negative real roots and conjugate complex roots. The Jordan block, composed of the opposites of the negative real roots, is used... (symbol Let m be the dimension of a matrix, i.e., the positive eigenvalue matrix; Jordan blocks, consisting of the opposites of the conjugate complex roots, are represented by... (symbol Let the dimension of a matrix be nm. These two types of Jordan blocks constitute the steady-state matrix. Jordan standard model :

[0070] (6)

[0071] in, For m rows A matrix in which all columns are zero. for A matrix with m rows and m columns of all zero elements serves to ensure that the Jordan canonical form is a square matrix. (Based on the Jordan canonical form...) Based on the properties of , we can obtain the following equation regarding real matrices:

[0072] (7)

[0073] in, It is a real matrix.

[0074] Step S4: Using auxiliary matrices, real matrices, and some equation transformations, derive the high dynamic output matrix, thereby constructing a minimum-phase dynamic system for controller design. The specific implementation process includes:

[0075] First, multiply the left side of equation (7) concerning real matrices by... We can obtain:

[0076] (8)

[0077] Then, the transpose of equation (8) is taken to obtain the following result:

[0078] (9)

[0079] Two auxiliary matrices are then defined as and (the symbol denotes that a matrix is composed of rows columns of elements):

[0080] (10)

[0081] where is an m-by-m identity matrix. Multiplying equation (9) by the matrix on both sides yields the following equation:

[0082] (11)

[0083] Since equation holds, equation (11) can be further expressed as:

[0084] (12)

[0085] Since , equation (12) can be further rewritten as:

[0086] (13)

[0087] where can be further simplified to the following form:

[0088] (14)

[0089] Substituting equation (14) into (13) yields:

[0090] (15)

[0091] Observing equation (15), it can be found that the positive eigenvalue matrix has appeared, which is a decisive factor affecting the high dynamic output matrix. Therefore, the high dynamic output matrix can be defined as , and equation (15) becomes the following form:

[0092] (16)

[0093] With the help of the high dynamic output matrix , the following output variable can be defined:

[0094] (17)

[0095] For output variables Taking the derivative, we get the following result:

[0096] (18)

[0097] in, for The derivative of . Finally, substituting the state-stable dynamic model (5) and equation (16) for the roll channel into (18), we can construct the minimum-phase dynamic system for controller design:

[0098] (19)

[0099] in For high dynamic roll angle stabilization control signals that require further design.

[0100] Step S5 introduces two key parameters: the positive eigenvalue matrix and the high dynamic output matrix, to further design the high dynamic roll angle stabilization control signal in the minimum phase dynamic system designed for the controller. The specific implementation process includes:

[0101] The entire high-dynamic roll angle stabilization control signal consists of two main parts: a proportional stabilization term and an integral stabilization term. The proportional stabilization term should include high-dynamic output matrix parameters, which can improve the roll channel state vector. The convergence speed. It is precisely because of... Includes roll angle ,therefore Will follow They converge quickly together. This invention incorporates a proportional stabilization term. Designed in the following form:

[0102] (20)

[0103] in The design parameter is greater than zero. However, driven by the proportional stabilization term with high dynamic characteristics, a large static error occurs in the roll angle as it approaches the equilibrium point. Therefore, to mitigate the impact of the static error on the convergence performance of the roll angle, the following integral stabilization term is further introduced:

[0104] (twenty one)

[0105] in, For a certain moment, the symbol Indicates from the initial time to the integral of the high dynamic roll angle stabilization control signal is designed as:

[0106] (22)

[0107] where the design parameter The smaller the design parameter is set, the greater the output value of the integral stabilization term is, and the stronger the static error suppression capability is. However, a larger integral stabilization term will slow down the convergence speed of the roll angle, so the design parameter should be as large as possible to achieve high dynamic convergence of the roll angle while meeting the given static error index.

[0108]

[0109] From the minimum phase dynamic system oriented to the controller design, it can be found that the high dynamic roll angle stabilization control signal satisfies the following equation relationship between the command roll moment and the gain matrix

[0110] (23)

[0111] where the roll channel moment of inertia is an inherent parameter of the unmanned aerial vehicle, and the moment of inertia measuring device can measure its accurate value. Next, the gain matrix is a gain matrix that needs to be set artificially, so it is also known. In addition, the roll channel state variable can be accurately fed back or estimated by a sensor, so it is also known. Then, the command roll moment can be inversely solved as follows by combining the expression (22) of the high dynamic roll angle stabilization control signal:

[0112] (24)

[0113] Step S7: The specific expression of the command roll moment is input into the coupled roll moment output characteristic model to obtain the specific value of the actual roll moment that should be provided for the roll channel dynamics of the unmanned aerial vehicle at each sampling time. The specific implementation process includes:

[0114] Substituting the equation into the coupled roll moment output characteristic model (1) can obtain:

[0115] (25)

[0116] ​Therefore, only need to put the designed command roll moment (24) into (25) to obtain the specific value of the actual roll moment that should be provided for the unmanned aerial vehicle roll channel dynamics at each sampling time.

[0117] As another specific embodiment of the present embodiment, a third-order linearized dynamics model based on coupled roll moment is constructed for the roll channel of the unmanned aerial vehicle. First, a kind of motor drive system carried by the unmanned aerial vehicle generates a coupled roll moment in the roll channel and the command roll moment satisfy the dynamic equation (coupled roll moment output characteristic model) as formula (1). Wherein, the bandwidth of the motor drive system , the moment of inertia of the roll channel .

[0118] Next, under the condition that , a third-order linearized dynamics model based on coupled roll moment can be obtained, as formula (4).

[0119] Wherein the roll channel state vector , the unstable state matrix and the constant control vector are respectively:

[0120] (26)

[0121] As another specific embodiment of the present embodiment, the pole placement technique is used to process the unstable state matrix existing in the third-order linearized dynamics model based on coupled roll moment, so as to construct a state stable dynamics model facing the roll channel.

[0122] The three poles configured for the unstable state matrix are respectively , , , wherein is a negative real number, and are a pair of conjugate complex roots. The output matrix parameters derived according to the conjugate complex roots are helpful to improve the high dynamic convergence characteristics of the roll angle. Then, the gain matrix is solved by using the and pole placement functions provided by MATLAB. By substituting the known constant control vector and the gain matrix into equation , the stable state matrix can be solved:

[0123] (27)

[0124] Wherein, The specific value of the dimension of Next, only need to introduce the data item in the third-order linearization dynamics model based on the coupling roll moment , and the state stability dynamics model facing the roll channel can be obtained, as shown in equation (5).

[0125] As another specific embodiment of the present embodiment, the Jordan standard form is constructed according to the eigenvalue components of the stable state matrix, so as to obtain the external eigenvalue matrix and the real matrix.

[0126] The reciprocal of the eigenvalue is 28, so the positive eigenvalue matrix , and the corresponding dimension variable . In addition, the reciprocal of the conjugate complex roots and is , respectively. Therefore, the Jordan block composed of this set of conjugate complex roots is expressed in the following form:

[0127] (28)

[0128] The above two types of Jordan block groups and compose the Jordan standard form of the stable state matrix :

[0129] (29)

[0130] Substitute the specific values of and into equation , and the real matrix can be solved:

[0131] (30)

[0132] Then define two auxiliary matrices and :

[0133] (31)

[0134] Therefore, combined with the specific values of the auxiliary matrix and the real matrix , the high dynamic output matrix is derived by using equation .

[0135] With the help of the high dynamic output matrix You can define output variables. , as in formula (17).

[0136] For output variables Differentiating the derivative yields formula (18).

[0137] in, for The derivative of . Finally, by substituting the state-stable dynamic model (5) and equation (16) for the roll channel into (18), the minimum phase dynamic system for controller design can be constructed, as shown in equation (19).

[0138] in, For high dynamic roll angle stabilization control signals that require further design.

[0139] As another specific embodiment of this example, two key parameters, the positive eigenvalue matrix and the high dynamic output matrix, are introduced to further design the high dynamic roll angle stabilization control signal in the minimum phase dynamic system designed for the controller.

[0140] The high dynamic roll angle stabilization control signal is designed as shown in formula (22). Wherein, the design parameters... , For a certain moment, the symbol Indicates from the initial time to The points.

[0141] In another specific embodiment of this example, the specific expression for the command roll torque is derived by inverse solving the high dynamic roll angle stabilization control signal. This is achieved using the equation... The command rolling torque, as shown in formula (24), can be obtained by inverse solution. As another specific embodiment of this example, the actual rolling torque at each sampling moment is calculated. The equation is then... Substituting into the coupled rolling torque output characteristic model yields formula (25). Therefore, only the designed command rolling torque needs to be applied. By substituting the values, we can obtain the actual torque that should be provided to the unmanned aerial vehicle's roll channel at each sampling time.

[0142] In this embodiment, comparative experiments are used to demonstrate that the proposed high-dynamic roll angle control method has faster convergence characteristics. For example, a proportional-integral controller (comparison controller) that can only be configured with negative real roots is also used to control the trajectory of the roll angle change, and its desired poles are configured sequentially as follows: , and The relevant experimental results are as follows: Figure 2 As shown. First, define a consistent boundary. To measure the dynamic characteristics of the controller. From Figure 2It can be found that the roll angle of the unmanned aerial vehicle needs 6.01 seconds to converge into the consistent boundary under the control of the comparative controller. However, under the action of the control method proposed in the application, the roll angle can be converged into the consistent boundary only in 3.90 seconds. Therefore, the dynamic characteristic of the control method is improved by about 2 times compared with the comparative controller.

[0143] The application derives a high dynamic output matrix by means of an auxiliary matrix, a real matrix and a series of equation transformations. The new construction method can configure conjugate complex roots for the state matrix, greatly improves the high dynamic convergence ability of the roll angle, and enables the unmanned aerial vehicle to realize the stabilization control of the roll angle more quickly and stably in a complex environment.

[0144] The application introduces an integral stabilization term with only one main control parameter, which can not only significantly reduce the control parameter adjustment period, but also improve the debugging efficiency of the control system. Meanwhile, the static error has always been an important factor affecting the convergence of the controller. The addition of the integral stabilization term effectively suppresses the adverse effects of the static error on the convergence of the controller, so that the controller is more stable and reliable during operation, can better cope with various disturbances and uncertainties, and thus improves the overall performance of the roll angle stabilization control of the unmanned aerial vehicle.

[0145] The above is only the preferred specific embodiment of the application, but the protection scope of the application is not limited thereto. Any changes or replacements easily thought of by those skilled in the art within the technical range disclosed in the application should be covered in the protection scope of the application. Therefore, the protection scope of the application should be subject to the protection scope of the claims.​

Claims

1. A method of stabilization control for a roll angle of an unmanned aerial vehicle, characterized by, The method comprises the following steps: constructing a third-order linearized dynamic model based on a roll channel of an unmanned aerial vehicle; processing an unstable state matrix in the third-order linearized dynamic model by using a pole placement technique to obtain a stable state matrix, and constructing a state stable dynamic model of the roll channel; constructing a Jordan canonical form based on eigenvalue components of the stable state matrix to obtain a positive eigenvalue matrix and a real number matrix; deriving a high dynamic output matrix by equation transformation based on the positive eigenvalue matrix, the real number matrix, and an auxiliary matrix; designing a high dynamic roll angle stabilization control signal based on the positive eigenvalue matrix and the high dynamic output matrix; solving the high dynamic roll angle stabilization control signal to obtain a command roll moment; inputting the command roll moment into a coupled roll moment output characteristic model to obtain an actual roll moment value of the unmanned aerial vehicle at each sampling time; The process of constructing a third-order linearized dynamic model based on the roll channel of the unmanned aerial vehicle comprises: establishing a coupled roll moment output characteristic model based on a coupled roll moment and a command roll moment of a motor drive system; constructing a linearized relationship among a roll angle, a roll angular velocity, and a roll moment; obtaining a third-order linearized dynamic model based on the coupled roll moment by combining the coupled roll moment output characteristic model and the linearized relationship through defining a roll channel state vector, an unstable state matrix, and a constant control vector; The expression of the coupled roll moment output characteristic model is: ； wherein is the derivative of is the bandwidth of the motor drive system, is the moment of inertia of the roll channel, is the inverse of is the coupled roll moment generated by the motor drive system carried by the unmanned aerial vehicle in the roll channel, is the commanded roll moment; The expression of the third-order linearized dynamic model based on the coupled roll moment is: ； In the formula, The matrix represents the unstable state. For constant control vectors, This is the roll channel state vector. for The derivative of for The reciprocal, For command rolling torque; , For the roll angular velocity, This is the roll angle.

2. The roll angle autopilot-oriented trim control method according to claim 1, wherein The process of constructing a Jordan canonical form based on eigenvalue components of the stable state matrix to obtain a positive eigenvalue matrix and a real number matrix comprises: classifying eigenvalues of the stable state matrix to obtain negative real roots and conjugate complex roots; constituting a positive eigenvalue matrix from reciprocals of the negative real roots; constituting a complex Jordan block from reciprocals of the conjugate complex roots; constructing a Jordan canonical form based on the positive eigenvalue matrix and the complex Jordan block; obtaining the real number matrix by solving a conversion relationship between the real number matrix and the Jordan canonical form.

3. The roll angle autopilot-oriented trim control method according to claim 2, wherein The process of deriving a high dynamic output matrix by equation transformation based on the positive eigenvalue matrix, the real number matrix, and an auxiliary matrix comprises: transposing an equation of the real number matrix and introducing an auxiliary matrix, and obtaining a simplified equation containing the positive eigenvalue matrix by eliminating intermediate variables through matrix operation; defining the high dynamic output matrix based on the simplified equation.

4. The roll angle autopilot-oriented trim control method according to claim 3, wherein The method further comprises constructing a minimum phase dynamic system for controller design based on the high dynamic output matrix and the state stable dynamic model of the roll channel; The expression of the minimum phase dynamic system for controller design is: ； wherein is the derivative of is the derivative of is the output variable, is the positive eigenvalue matrix, T is the transpose, is the high dynamic output matrix, is the constant control vector, is the high dynamic roll angle stabilization control signal which needs further design.

5. The roll angle autopilot-oriented trim control method according to claim 4, wherein The expression of the designed high dynamic roll angle stabilization control signal is: ； wherein is a high dynamic roll angle stabilizing control signal that needs further design, is a proportional stabilizing term, is an integral stabilizing term, T is transpose, is a high dynamic output matrix, is a certain time instant, is a design parameter greater than zero, is a roll channel state vector, is a positive eigenvalue matrix, denotes the integral from the initial time instant to is a constant control vector.​ 6. A computer comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the roll angle stabilization control method for unmanned aerial vehicles in claim 1.

7. A storage medium having stored thereon a computer program, characterized in that The program is executed by the processor to implement the roll angle stabilization control method for unmanned aerial vehicles in claim 1.

Images