Unmanned helicopter compound model predictive control method considering preset performance

Abstract

The application discloses a kind of unmanned helicopter compound model predictive control methods considering preset performance, first, the disturbed dynamic model of unmanned helicopter affected by interference is established, and the trajectory tracking problem is converted into the instruction tracking problem of position and attitude loop;Second, the estimation of lumped disturbance of position loop and attitude loop is realized by designing finite-time disturbance observer, and the dynamic and steady-state constraints of tracking error are realized based on position tracking error design performance function;Finally, combined with disturbance estimation information and nonlinear model predictive control algorithm, compound preset performance nonlinear model predictive controller and compound predictive controller are designed in position and attitude loop respectively.The application realizes the direct compensation of the lumped disturbance of unmanned helicopter, while meeting the tracking error constraint, improves the trajectory tracking accuracy and flight safety of unmanned helicopter in the environment of interference and actuator fault, and ensures that the trajectory tracking error convergence process is always within the preset envelope.

Description

Technical Field

[0001] This invention belongs to the field of flight control technology, specifically relating to a composite model predictive control method for unmanned helicopters that considers preset performance. Background Technology

[0002] Unmanned helicopters (UHHs) have gained widespread application in both military and civilian fields due to their unique vertical takeoff and landing, hovering, and low-altitude flight capabilities. Since they operate without a pilot, UHHs execute flight missions by automatically tracking pre-designed trajectories, making trajectory tracking control a core task in UHH control system design. As flight missions become increasingly complex, the flight environments in which UHHs operate are also becoming more complex, inevitably subjecting them to multi-source interference from external environmental disturbances, internal parameter perturbations, and unmodeled dynamics. Furthermore, the high complexity of the transmission and main rotor structure makes UHHs highly vulnerable to actuator failures. High-precision trajectory tracking control under conditions of multi-source interference and actuator failures has become a significant challenge in the design of UHH control systems.

[0003] To address the aforementioned issues, scholars both domestically and internationally have proposed various advanced control schemes, including adaptive control, sliding mode control, and backstepping control. Most of these schemes rely on error feedback signals to passively suppress disturbances and rarely simultaneously consider transient performance constraints and fault tolerance requirements. Model predictive control can handle constraints, but it is sensitive to disturbances; disturbance observers can actively estimate and compensate for disturbances, but lack guarantees for the transient performance of tracking errors; preset performance control can ensure that the tracking error convergence process meets a preset envelope, but existing preset performance control schemes are mostly combined with backstepping methods, resulting in computational complexity and weak constraint handling capabilities. Therefore, there is an urgent need to propose a composite control method for unmanned helicopters that can achieve rapid compensation for multi-source disturbances and faults, meet preset performance constraints, and is easy to implement in engineering. Summary of the Invention

[0004] Purpose of the invention: In order to overcome the shortcomings of the prior art, this invention proposes a composite model prediction fault-tolerant control method for unmanned helicopters that considers preset performance. This method achieves accurate estimation and compensation of lumped disturbances within a finite time, ensuring that the trajectory tracking error convergence process of the unmanned helicopter is always within the preset envelope, thus possessing high-precision trajectory tracking capability. This effectively solves the problem of high-precision and high-safety trajectory tracking control of unmanned helicopters under disturbances and actuator failures.

[0005] Technical solution: The present invention provides a composite model predictive control method for unmanned helicopters that considers preset performance, specifically including the following steps:

[0006] S1. Establish a full-loop disturbed dynamic model of the unmanned helicopter, including the effects of external disturbances and actuator failures, and construct dynamic equations for the tracking errors of the position subsystem and attitude subsystem.

[0007] S2. Design finite-time disturbance observers for the position tracking error system and the attitude tracking error system respectively, so as to realize finite-time estimation of the lumped disturbance of the position loop and the lumped disturbance of the attitude loop.

[0008] S3. Based on the position tracking error constraints, design a preset performance function and construct a dynamic equation for the transformation error;

[0009] S4. Based on the nonlinear model predictive control algorithm, and combined with the estimation information of the lumped disturbance of the position loop by the finite-time disturbance observer, design a nonlinear model predictive controller with composite preset performance for the position loop, generate virtual control quantities for the position loop, and solve the main rotor lift command value, roll angle command value and pitch angle command value based on the virtual control quantities.

[0010] S5. Based on the nonlinear model predictive control algorithm, and combined with the estimation information of the lumped disturbance of the attitude loop by the finite-time disturbance observer, a composite nonlinear model predictive controller for the attitude loop is designed to generate the virtual control quantity of the attitude loop, and the real control quantity of the unmanned helicopter is generated by inverse solution.

[0011] Furthermore, in step S1, establishing the full-loop disturbed dynamics model of the unmanned helicopter specifically involves:

[0012] Establish the position dynamic equations for the unmanned helicopter:

[0013] (1)

[0014] Establish the attitude dynamic equations of the unmanned helicopter:

[0015] (2)

[0016] in, These represent the position and velocity of the unmanned helicopter in the inertial coordinate system, respectively. and These represent the position coordinates and velocity components along the three coordinate axes in the inertial frame. The acceleration due to gravity is constant. For quality, A three-dimensional vector in an inertial frame. Main rotor lift, The external disturbance force experienced by the position loop; The attitude angle of the unmanned helicopter. These represent roll angle, pitch angle, and yaw angle, respectively. This is the transformation matrix from the body's axisymmetric frame to its inertial frame. This is the transformation matrix between the rate of change of attitude angle and the attitude angular velocity under the machine system.

[0017] Furthermore, in step S1, constructing the dynamic equations for the tracking errors of the position subsystem and attitude subsystem specifically involves:

[0018] Main rotor lift of unmanned helicopter and the three-axis torque of unmanned helicopters The calculation is based on the control quantities of the unmanned helicopter, which include... , representing the collective pitch, longitudinal cyclic pitch, lateral cyclic pitch, and tail rotor pitch of the unmanned helicopter, respectively; the specific calculation is shown below:

[0019] (3)

[0020] in,

[0021]

[0022] in, These are constant parameters, all related to the structure of the unmanned helicopter. The total distance failure factor, These are the longitudinal and lateral periodic distance fault coefficients, respectively. The tail rotor pitch failure factor. The velocity of the unmanned helicopter along the Z-axis of its fuselage;

[0023] Introducing virtual control quantity in the position loop :

[0024]

[0025] in, The nominal value of the main rotor lift; the position tracking error along the X, Y, and Z axes in the inertial frame is defined. , The position loop command given for trajectory tracking is given by equation (1). Based on the definition of virtual control quantity, the dynamic equation of the tracking error of the position subsystem is obtained:

[0026] (4)

[0027] Based on the definition of virtual control variables, the command value of the main rotor lift corresponding to the virtual control variables can be obtained through inverse kinematics. and roll angle command value and pitch angle command value :

[0028] (5)

[0029] Define attitude angle tracking error ,in, The yaw angle command given for trajectory tracking, These are the roll angle and pitch angle commands calculated from the position loop trajectory tracking control, respectively.

[0030] Combining equation (2), the dynamic equation of the attitude subsystem tracking error is constructed by calculating the difference between the actual attitude angle and the desired attitude angle:

[0031] (6)

[0032] Introduction , ,in, Virtual control variables introduced for the attitude control subsystem of unmanned helicopters. This is a non-linear term, specifically as follows:

[0033]

[0034] The dynamic equation for the tracking error of the attitude subsystem is simplified to obtain:

[0035] (7).

[0036] Furthermore, in step S2, the finite-time interference observer designed for the position tracking error system is as follows:

[0037] X-channel: (8)

[0038] Y-channel: (9)

[0039] Z-channel: (10)

[0040] in, Let be the observer parameters, whose values ​​satisfy... ; This is an estimate of the lumped interference across the three channels.

[0041] Furthermore, in step S2, the finite-time disturbance observer designed for the attitude tracking error system is as follows:

[0042] Roll angle channel: (11)

[0043] Pitch Angle Channel: (12)

[0044] Yaw angle channel: (13)

[0045] in, Let be the observer parameters, whose values ​​satisfy... ; These are estimated values ​​of the derivatives of the tracking errors for the roll, pitch, and yaw angles, respectively. These are the estimated values ​​for the lumped interference in the roll, pitch, and yaw channels, respectively.

[0046] Furthermore, the specific implementation process of step S3 is as follows:

[0047] Introducing a performance function that converges asymptotically over time to constrain the tracking error, for the X channel, the constraint function is:

[0048] (14)

[0049] in, For performance functions, The required positive constants are those that satisfy the following conditions. ;

[0050] The inequality constraint on tracking error is transformed into an equality constraint, and variables are introduced. and define about Transformation error variable Then, by inverse solution, we can obtain... about Functions:

[0051] (15)

[0052] if only Boundedness guarantees It is valid; and both have the same monotonicity, which can be achieved through control design. Approaching zero, thus ensuring tracking error Approaching zero;

[0053] Select the X channel according to equation (14) Make Within the preset performance function envelope, we have:

[0054]

[0055] Combining the position loop tracking error dynamics, the position loop X channel conversion error can be obtained by simultaneously solving equations (14) and (15). Updates:

[0056] (16)

[0057] in, and For the X channel regarding preset performance status , , The function; conversion error The dynamics are determined by the control quantity Decision-making, control ensure Bounded, ensuring tracking error Within the preset envelope.

[0058] Furthermore, the position loop composite preset performance nonlinear model predictive controller mentioned in step S4 is:

[0059] (18)

[0060] in, These are functions of the three channels of the position ring with respect to the preset performance state; These are the prediction step sizes for the three channels of the position ring.

[0061] Furthermore, the attitude loop composite nonlinear model predictive controller described in step S5 is:

[0062] (19)

[0063] in, These are the estimated values ​​of the lumped interference in the three channels of the attitude loop. These are the prediction step sizes for the three channels of the attitude loop.

[0064] Furthermore, the actual control parameters of the unmanned helicopter system described in step S5 are:

[0065] (20)

[0066] in, These represent the collective pitch, longitudinal cyclic pitch, lateral cyclic pitch, and tail rotor pitch of the unmanned helicopter, respectively.

[0067] Beneficial effects: Compared with the prior art, the present invention has the following beneficial effects: The present invention achieves accurate estimation of lumped interference and attitude angle tracking error change rate within a finite time, as well as high-precision trajectory tracking under multi-source interference and rapid fault tolerance under fault action; The present invention improves the trajectory tracking accuracy and ensures that the trajectory tracking error convergence process is always within the preset envelope, effectively improving flight safety. Attached Figure Description

[0068] Figure 1 This is a flowchart of the present invention;

[0069] Figure 2 This is a three-dimensional position tracking curve of the proposed method based on a composite pre-defined performance nonlinear model predictive controller (PPC+CNMPC-FTDO) with a finite-time disturbance observer and a comparison method.

[0070] Figure 3These are error response curves for position three-axis trajectory tracking under different methods;

[0071] Figure 4 These are attitude angle command tracking error curves under different methods;

[0072] Figure 5 These are the response curves of the actual control quantity under different methods;

[0073] Figure 6 This is the location loop lumped disturbance estimation response curve;

[0074] Figure 7 It is a curve of the three-axis virtual control quantity of the position under the composite method. Detailed Implementation

[0075] The present invention will now be described in further detail with reference to the accompanying drawings.

[0076] like Figure 1 As shown, the present invention provides a composite model predictive control method for unmanned helicopters that considers preset performance, and the specific implementation process is as follows:

[0077] S1. Establish a full-loop disturbed dynamic model of the unmanned helicopter that includes the effects of external disturbances and actuator failures, and construct dynamic equations for the tracking errors of the position and attitude subsystems.

[0078] Establish the position dynamic equations for the unmanned helicopter:

[0079] (1)

[0080] Establish the attitude dynamic equations of the unmanned helicopter:

[0081] (2)

[0082] in, These represent the position and velocity of the unmanned helicopter in the inertial coordinate system, respectively. and These represent the position coordinates and velocity components along the three coordinate axes in the inertial frame. The acceleration due to gravity is constant. For quality, A three-dimensional vector in an inertial frame. Main rotor lift, The external disturbance force experienced by the position loop; The attitude angle of the unmanned helicopter. These represent roll angle, pitch angle, and yaw angle, respectively. This is the transformation matrix from the body's axisymmetric frame to its inertial frame, specifically in the form of:

[0083]

[0084] in, Let ω be the angular velocity about the three axes of the body coordinate system. These are roll rate, pitch rate, and yaw rate, respectively. The transformation matrix between the rate of change of attitude angle and the attitude angular velocity under the machine system is as follows:

[0085]

[0086] in, Here is the rotational inertia matrix. for The inverse matrix, For aerodynamic torque, These are rolling torque, pitching moment, and yaw moment, respectively. This represents the external disturbance torque experienced by the attitude loop.

[0087] In this implementation case, the initial state of the unmanned helicopter is set as follows:

[0088]

[0089] Main rotor lift of unmanned helicopter and the three-axis torque of unmanned helicopters The calculation is based on the control quantities of the unmanned helicopter, which include... , representing the collective pitch, longitudinal cyclic pitch, lateral cyclic pitch, and tail rotor pitch of the unmanned helicopter, respectively; the specific calculation is shown below:

[0090] (3)

[0091] in,

[0092]

[0093] in, These are constant parameters, all related to the structure of the unmanned helicopter. The total distance failure factor, These are the longitudinal and lateral periodic distance fault coefficients, respectively. The tail rotor pitch failure factor. The velocity of the unmanned helicopter along the Z-axis of its fuselage.

[0094] Define the position tracking error along the X, Y, and Z axes in the inertial frame. ,

[0095] in, These are the position loop commands given for trajectory tracking.

[0096] The command to set the three-dimensional reference trajectory of the unmanned helicopter's position as a function of time t is as follows:

[0097]

[0098] The dynamic equation for the tracking error of the position subsystem is constructed by calculating the difference between the actual position and the desired position of the unmanned helicopter:

[0099] (4)

[0100] in, The virtual control variables introduced for the unmanned helicopter position control subsystem are as follows:

[0101]

[0102] The command value of the main rotor lift can be obtained by reversing the above formula. and roll angle command value and pitch angle command value :

[0103] (5)

[0104] Define attitude angle tracking error ,

[0105] in, The given yaw angle command for trajectory tracking, These are the roll angle command and pitch angle command calculated from the position loop trajectory tracking control, respectively.

[0106] The dynamic equation for the tracking error of the attitude subsystem is constructed by calculating the difference between the actual attitude angle and the desired attitude angle:

[0107] (6)

[0108] For ease of calculation, we introduce... , ,in, Virtual control variables introduced for the attitude control subsystem of unmanned helicopters. This is a non-linear term, specifically as follows:

[0109]

[0110] The dynamic equation for the tracking error of the attitude subsystem is simplified to obtain:

[0111] (7)

[0112] The model parameters of the unmanned helicopter are shown in Table 1.

[0113] Table 1 Parameter List

[0114]

[0115] S2. Design finite-time disturbance observers for the position and attitude tracking error systems respectively to achieve finite-time estimation of lumped disturbances.

[0116] Design a finite-time disturbance observer for the position loop under position error conditions:

[0117] X-channel: (8)

[0118] Y-channel: (9)

[0119] Z-channel: (10)

[0120] in, For observer parameters, ; This is an estimate of the lumped interference across the three channels.

[0121] Design a finite-time disturbance observer for the attitude loop in response to attitude error states:

[0122] Roll angle channel: (11)

[0123] Pitch Angle Channel: (12)

[0124] Yaw angle channel: (13)

[0125] in, For observer parameters, ; These are estimated values ​​of the derivatives of the tracking errors for the roll, pitch, and yaw angles, respectively. These are the estimated values ​​for the lumped interference in the roll, pitch, and yaw channels, respectively.

[0126] S3. Combining position tracking error constraints, design a preset performance function and construct a dynamic equation for the transformation error.

[0127] To ensure that the position tracking error convergence process satisfies the constraints, a performance function that converges asymptotically over time is introduced to constrain the tracking error. Taking the X channel as an example, the constraint function is given as follows:

[0128] (14)

[0129] in, For the performance function, select The Y and Z channels are similar, with the following parameters selected:

[0130]

[0131] To facilitate controller design, the inequality constraints of tracking error are transformed into equality constraints, and variables are introduced. and define about Transformation error variable Then it can be solved inversely to obtain about Functions:

[0132] (15)

[0133] if only Boundedness guarantees It is valid; and both have the same monotonicity, which can be achieved through control design. Approaching zero, thus ensuring tracking error It approaches zero.

[0134] Select a suitable X channel according to equation (14) Make Within the preset performance function envelope, we have:

[0135]

[0136] Combining the position loop tracking error dynamics, the position loop X channel conversion error can be obtained by simultaneously solving equations (14) and (15). Updates:

[0137] (16)

[0138] in, .

[0139] The conversion error can be seen from equation (16). The dynamics are determined by the control quantity It is determined, so as long as the quantity is controlled. ensure Boundedness ensures tracking error Within the preset envelope.

[0140] S4. Based on the model predictive control algorithm and combined with disturbance estimation information, a composite model predictive controller with preset performance for position loop is designed.

[0141] Taking the X channel as an example, based on the model predictive control method and combined with the dynamic equation of the tracking error of the position subsystem, considering the influence of disturbances, the composite preset performance nonlinear model predictive controller of the X channel in the position loop is obtained:

[0142] (17)

[0143] The other channels of the position loop are similar to those described above, resulting in a position loop composite preset performance nonlinear model predictive controller:

[0144] (18)

[0145] in, .

[0146] S5. Combining attitude loop estimation information, design an attitude loop composite model predictive controller, and generate the actual control variables of the unmanned helicopter through inverse kinematics.

[0147] Similar to the position loop, the attitude loop, combined with the dynamic equation of the tracking error, and considering the influence of disturbances, yields a composite pre-defined performance nonlinear model predictive controller for the attitude loop:

[0148] (19)

[0149] in, ;

[0150] By combining equations (3), (18), and (19), the control variables of the unmanned helicopter system are obtained. :

[0151] (20)

[0152] This invention utilizes the Simulink toolbox in the simulation software MATLAB R2025b to build a full-loop dynamic model of an unmanned helicopter containing lumped disturbances, and then conducts simulations and experiments.

[0153] Position loop and attitude loop interference settings are as follows:

[0154]

[0155] in, `<value>` is a non-negative integer representing the piecewise function expression corresponding to different time periods. An executor fault is injected at 10 seconds.

[0156] .

[0157] Figures 2 to 7The paper presents the three-dimensional position tracking, three-axis trajectory tracking error response, yaw angle command tracking error, actual control quantity response, position loop lumped disturbance estimation, and position three-axis virtual control quantity response curves under the action of the proposed method: a composite model predictive safety controller (PPC+CNMPC-FTDO), a baseline nonlinear model predictive control (BNMPC), a baseline nonlinear model predictive control with prescribed performance (PPC+BNMPC), and a composite nonlinear model predictive control with prescribed performance based on an extended state observer (PPC+CNMPC-ESO). Figure 2 It can be seen that BNMPC's tracking performance is poor, while the other three schemes, which consider preset performance, can all guarantee effective tracking of the 3D trajectory to a certain extent. From Figure 3 It can be seen that the scheme considering the preset performance can ensure that the three-axis position tracking error always falls within the preset envelope, thereby ensuring that the transient and steady-state of the unmanned helicopter meets the preset constraints. The composite predictive control scheme including interference observation compensation has higher tracking accuracy. The composite predictive control scheme based on FTDO proposed in this invention has the highest trajectory tracking accuracy. Figure 4 The figure shows the attitude angle tracking error response curve of the unmanned helicopter. The proposed scheme PPC+CNMPC-FTDO outperforms the other three control schemes in both transient and steady-state response. Figure 5 The actual control response curves of the unmanned helicopter under four control schemes are presented. It can be seen that the control amplitudes are similar under the four schemes, and the required energy is on the same order of magnitude, which shows the fairness of the experimental comparison. Figure 6 The position loop lumped disturbance estimation response curve is shown. It can be seen that the FTDO used in the proposed scheme achieves higher disturbance estimation accuracy compared to the ESO in the comparative scheme, thus realizing higher-precision disturbance feedforward compensation and ultimately improving the tracking accuracy of the unmanned helicopter's 3D trajectory. Figure 7 As shown, all different methods satisfy the control quantity condition.

[0158] In summary, compared to the baseline nonlinear model predictive control, this invention exhibits superior anti-interference and constraint satisfaction characteristics. While satisfying physical limits, the finite-time disturbance observer achieves high-precision estimation of lumped disturbances, ensuring the steady-state accuracy of trajectory tracking. Simultaneously, the introduction of a preset performance function ensures that the convergence process of the position tracking error always remains within a pre-defined envelope, effectively guaranteeing the transient performance and flight safety of the unmanned helicopter.

[0159] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A composite model predictive control method for unmanned helicopters considering preset performance, characterized in that, Includes the following steps: S1. Establish a full-loop disturbed dynamic model of the unmanned helicopter, including the effects of external disturbances and actuator failures, and construct dynamic equations for the tracking errors of the position subsystem and attitude subsystem. S2. Design finite-time disturbance observers for the position tracking error system and the attitude tracking error system respectively, and realize finite-time estimation of the lumped disturbance of the position loop and the lumped disturbance of the attitude loop. S3. Based on the position tracking error constraints, design a preset performance function and construct a dynamic equation for the transformation error; S4. Based on the nonlinear model predictive control algorithm, and combined with the estimation information of the lumped disturbance of the position loop by the finite-time disturbance observer, design a nonlinear model predictive controller with composite preset performance for the position loop, generate virtual control quantities for the position loop, and solve the main rotor lift command value, roll angle command value and pitch angle command value based on the virtual control quantities. S5. Based on the nonlinear model predictive control algorithm, and combined with the estimation information of the lumped disturbance of the attitude loop by the finite-time disturbance observer, a composite nonlinear model predictive controller for the attitude loop is designed to generate the virtual control quantity of the attitude loop, and the real control quantity of the unmanned helicopter is generated by inverse solution.

2. The composite model predictive control method for unmanned helicopters considering preset performance as described in claim 1, characterized in that, In step S1, establishing the full-loop disturbed dynamics model of the unmanned helicopter specifically involves: Establish the position dynamic equations for the unmanned helicopter: (1) Establish the attitude dynamic equations of the unmanned helicopter: (2) in, These represent the position and velocity of the unmanned helicopter in the inertial coordinate system, respectively. and These represent the position coordinates and velocity components along the three coordinate axes in the inertial frame. The acceleration due to gravity is constant. For quality, A three-dimensional vector in an inertial frame. Main rotor lift, The external disturbance force experienced by the position loop; The attitude angle of the unmanned helicopter. These represent roll angle, pitch angle, and yaw angle, respectively. This is the transformation matrix from the body's axisymmetric frame to its inertial frame. This is the transformation matrix between the rate of change of attitude angle and the attitude angular velocity under the machine system.

3. The composite model predictive control method for unmanned helicopters considering preset performance, as described in claim 1, is characterized in that... In step S1, the dynamic equations for the tracking errors of the position subsystem and attitude subsystem are constructed as follows: Main rotor lift of unmanned helicopter and the three-axis torque of unmanned helicopters The calculation is based on the control quantities of the unmanned helicopter, which include... , representing the collective pitch, longitudinal cyclic pitch, lateral cyclic pitch, and tail rotor pitch of the unmanned helicopter, respectively; the specific calculation is shown below: (3) in, in, These are constant parameters, all related to the structure of the unmanned helicopter. The total distance failure factor, These are the longitudinal and lateral periodic distance fault coefficients, respectively. The tail rotor pitch failure factor. The velocity of the unmanned helicopter along the Z-axis of its fuselage; Introducing virtual control quantity in the position loop : in, The nominal value of the main rotor lift; the position tracking error along the X, Y, and Z axes in the inertial frame is defined. , The position loop command given for trajectory tracking is given by equation (1). Based on the definition of virtual control quantity, the dynamic equation of the tracking error of the position subsystem is obtained: (4) Based on the definition of virtual control variables, the command value of the main rotor lift corresponding to the virtual control variables can be obtained through inverse kinematics. and roll angle command value and pitch angle command value : (5) Define attitude angle tracking error ,in, The yaw angle command given for trajectory tracking, These are the roll angle and pitch angle commands calculated from the position loop trajectory tracking control, respectively. Combining equation (2), the dynamic equation of the attitude subsystem tracking error is constructed by calculating the difference between the actual attitude angle and the desired attitude angle: (6) Introduction , ,in, Virtual control variables introduced for the attitude control subsystem of unmanned helicopters. This is a non-linear term, specifically as follows: The dynamic equation for the tracking error of the attitude subsystem is simplified to obtain: (7)。 4. The composite model predictive control method for unmanned helicopters considering preset performance as described in claim 1, characterized in that, In step S2, the finite-time interference observer designed for the position tracking error system is as follows: X-channel: (8) Y-channel: (9) Z-channel: (10) in, Let be the observer parameters, whose values ​​satisfy... ; This is an estimate of the lumped interference across the three channels.

5. The composite model predictive control method for unmanned helicopters considering preset performance as described in claim 1, characterized in that, In step S2, the finite-time disturbance observer for the attitude tracking error system is designed as follows: Roll angle channel: (11) Pitch Angle Channel: (12) Yaw angle channel: (13) in, Let be the observer parameters, whose values ​​satisfy... ; These are estimated values ​​of the derivatives of the tracking errors for the roll, pitch, and yaw angles, respectively. These are the estimated values ​​for the lumped interference in the roll, pitch, and yaw channels, respectively.

6. The composite model predictive control method for unmanned helicopters considering preset performance as described in claim 1, characterized in that, The specific implementation process of step S3 is as follows: Introducing a performance function that converges asymptotically over time to constrain the tracking error, for the X channel, the constraint function is: (14) in, For performance functions, The required positive constants are those that satisfy the following conditions. ; The inequality constraint on tracking error is transformed into an equality constraint, and variables are introduced. and define about Transformation error variable Then, by inverse solution, we can obtain... about Functions: (15) if only Boundedness guarantees It is valid; and both have the same monotonicity, which can be achieved through control design. Approaching zero, thus ensuring tracking error Approaching zero; Select the X channel according to equation (14) Make Within the preset performance function envelope, we have: Combining the position loop tracking error dynamics, the position loop X channel conversion error can be obtained by simultaneously solving equations (14) and (15). Updates: (16) in, and For the X channel regarding preset performance status , , The function; conversion error The dynamics are determined by the control quantity Decision-making, control ensure Bounded, ensuring tracking error Within the preset envelope.

7. The composite model predictive control method for unmanned helicopters considering preset performance as described in claim 1, characterized in that, The position loop composite preset performance nonlinear model predictive controller mentioned in step S4 is: (18) in, These are functions of the three channels of the position ring with respect to the preset performance state; These are the prediction step sizes for the three channels of the position ring.

8. The predictive control method for a composite model of an unmanned helicopter considering preset performance, as described in claim 1, is characterized in that, The attitude loop composite nonlinear model predictive controller mentioned in step S5 is: (19) in, These are the estimated values ​​of the lumped interference in the three channels of the attitude loop. These are the prediction step sizes for the three channels of the attitude loop.

9. The composite model predictive control method for unmanned helicopters considering preset performance as described in claim 1, characterized in that, The actual control parameters of the unmanned helicopter system described in step S5 are: (20) in, These represent the collective pitch, longitudinal cyclic pitch, lateral cyclic pitch, and tail rotor pitch of the unmanned helicopter, respectively.

Images