A constrained active disturbance rejection fault-tolerant control method for unmanned helicopters

Abstract

The present invention provides an active anti-disturbance fault-tolerant control method for an unmanned helicopter taking constraints into consideration. First, a disturbed nonlinear dynamic model of the attitude loop of the unmanned helicopter including actuator failures is established. Secondly, based on the extended state observer technology, an effective estimation of external interference is performed. Then, a composite anti-interference controller for the attitude loop is designed by combining the interference estimation information and a dynamic gain constraint processing control algorithm to obtain the required force and torque. Finally, the helicopter's cyclic pitch and tail rotor pitch are solved based on the required force and torque. The present invention fully considers the limiting requirements of the unmanned helicopter's attitude angle change rate, and performs feedforward compensation for interference based on the idea of ​​active anti-interference control. Under the condition of satisfying the constraints, high-precision tracking of the attitude angle command of the unmanned helicopter is achieved, effectively reducing the impact of multi-source interference and system failures.

Description

Technical Field

[0001] The invention belongs to the technical field of flight control, and in particular relates to an active anti-disturbance fault-tolerant control method for an unmanned helicopter considering constraints. Background Art

[0002] Unmanned helicopters are unmanned rotorcraft. They have attracted widespread attention due to their low cost, high flexibility, and ability to take off and land vertically. They are widely used in both military and civilian fields. The three-axis acceleration of a helicopter is mainly controlled by the lift of the main rotor decomposed into the component forces of each axis through the change of the attitude angle. Therefore, stable attitude control is the prerequisite for full-loop control design. The helicopter attitude system has the characteristics of strong nonlinearity and strong coupling. In addition, it is susceptible to multi-source interference such as external gusts, internal parameter perturbations, and unmodeled dynamics during flight, as well as the threat of actuator failure. These characteristics pose great challenges to control system design. Rapidly changing attitude angles can reduce flight stability and even threaten flight safety. Therefore, the attitude angle change rate constraint problem has received widespread attention in academia and industry.

[0003] In response to the influence of multi-source interference, existing methods usually rely on the robustness of the algorithm to indirectly suppress multi-source interference or use active anti-interference control methods based on observers to achieve rapid compensation of interference. However, there is no method that handles attitude angular rate constraints while considering the influence of interference. Summary of the invention

[0004] Purpose of the invention: The invention aims at the attitude system of unmanned helicopters affected by attitude angle change rate constraints, actuator failures and multi-source interference. Based on the extended state observer technology and dynamic gain constraint processing method, the invention provides an unmanned helicopter active anti-disturbance fault-tolerant control method with constraint processing capability, which ensures the high-precision tracking of attitude angle commands of unmanned helicopters under the condition of meeting approximately. By estimating and compensating for the lumped interference, the influence of multi-source interference and failures is effectively suppressed, and the tracking accuracy of the attitude angle is improved.

[0005] The present invention specifically provides a method for active anti-disturbance fault-tolerant control of an unmanned helicopter considering constraints, comprising the following steps:

[0006] Step 1, establishing a disturbed nonlinear dynamic model of the attitude loop of an unmanned helicopter including actuator failure;

[0007] Step 2: Based on the extended state observer technique, an interference observer is designed to estimate the aggregate interference.

[0008] Step 3, designing a composite anti-interference controller for the attitude loop by combining the interference estimation information and the dynamic gain constraint processing control method;

[0009] Step 4, based on the required torque, the actual control variables of the helicopter, cyclic pitch and tail rotor pitch, are obtained by inverse solution.

[0010] Step 1 contains the following steps:

[0011] Step 1-1, establishing a posture loop dynamics model affected by multiple sources of interference;

[0012] Step 1-2, establish the actuator model including the fault effect.

[0013] In step 1-1, the attitude loop dynamics model affected by multi-source interference is as follows:

[0014]

[0015] where Θ = [φ,θ,ψ] T is the helicopter attitude angle, φ, θ, ψ are roll angle, pitch angle, and yaw angle respectively. is the column vector of the first-order derivative of the helicopter attitude angle with respect to time, They are the first-order derivative of the roll angle with respect to time, the first-order derivative of the pitch angle with respect to time, and the first-order derivative of the yaw angle with respect to time; Ω = [p, q, r] T is the angular velocity of rotation around the body axis, p, q, r are roll angle, pitch angle, and yaw rate respectively; J = diag (I x ,I y ,I z ) is the moment of inertia matrix, I x ,I y ,I z are the moments of inertia around the three axes of the fuselage; Γ=[L,M,N] T is the three-axis moment, L, M, N are rolling moment, pitch moment and yaw moment respectively, D A is the multi-source interference received by the attitude loop, which is a vector with three rows and one column; T(Θ) is the transformation matrix, and its specific expression is:

[0016]

[0017] Among them, sinφ=sφ, cosφ=cφ, tanθ=tθ, sinθ=sθ, cosθ=cθ.

[0018] Steps 1-2 include:

[0019] The relationship between the rolling torque L, the pitching moment M and the periodic moment is as follows:

[0020]

[0021] Where L a,L b ,M a ,M b ,A lon ,A lat ,B lon ,B lat are system parameters, ε, τ are time constants, are the failure coefficients of the longitudinal period distance and the transverse period distance, respectively, δ lon ,δ lat are the longitudinal periodic distance and the transverse periodic distance respectively;

[0022] Introduce the intermediate parameter A defined as follows 0 , A 1 ,Φ,

[0023]

[0024] Then equation (2) can be rewritten as:

[0025]

[0026] Among them, L n ,M n They are respectively the nominal rolling torque and pitching moment when the period distance is fault-free, L f ,M f They are respectively the rolling torque and pitching moment caused by cyclic pitch fault, as shown below:

[0027]

[0028] The relationship between the yaw moment, tail rotor pitch and total pitch is as follows:

[0029]

[0030] Where N r ,N col ,N ped is the system parameter, δ col is the main rotor pitch, is the main rotor pitch failure coefficient, δ ped is the tail rotor pitch, is the tail rotor pitch failure coefficient, N n is the nominal value of the yaw moment without considering the fault, N f The yaw moment caused by tail rotor failure is as follows:

[0031] N n =N r r+N col δ col +N ped δ ped ,

[0032] Step 2 includes: defining the attitude angle tracking error:

[0033]

[0034] Where e Θ is the attitude angle tracking error column vector, Θ is the attitude angle vector, Θ d =[φ d ,θ d ,ψ r ] T is a three-row and one-column attitude angle command vector, φ d ,θ d ,ψ r They are roll angle command, pitch angle command and yaw angle command respectively;

[0035] According to equation (2), the tracking error dynamics of the attitude system is obtained:

[0036]

[0037] in For e Θ The column vector of the second-order derivatives of each element with respect to time, Γ n is the nominal value of the torque around the fuselage axis when there is no fault in the cyclic pitch, collective pitch and tail pitch, L n ,M n ,N n They are the nominal values ​​of rolling torque, pitching moment and yaw moment respectively:

[0038] Γ n =[L n M n N n ] T

[0039] is the matrix after the derivative of each element of the T(Θ) matrix with respect to t, as follows:

[0040]

[0041] Among them, Γ f is the torque variation caused by the total pitch, tail rotor pitch and periodic pitch change faults, specifically L f ,M f ,N f They are respectively the change of rolling moment, the change of pitch moment and the change of yaw moment; D is the column vector of the second-order derivative of the attitude angle command vector with respect to time. LA It is a vector with three rows and one column, which is the lumped disturbance, including the information of disturbance, fault and the second-order derivative of attitude angle command:

[0042] Γ f =[L f M f N f ] T ,

[0043] For formula (5), the interference observer is designed to estimate the aggregate interference as follows:

[0044]

[0045] Where z 1 ,z 2 ,z 3 is the state variable of the observer, For z 1 ,z 2 ,z 3 The first-order derivative with respect to time, the observer gain matrix For R 3×3 The diagonal matrix of is an estimate of the aggregate interference.

[0046] Step 3 includes:

[0047] For formula (5), the following definition is introduced:

[0048]

[0049] where τ 1 ,τ 2 ,τ 3 are virtual rolling moment, virtual pitch moment and virtual yaw moment, respectively, 1 ,f 2 ,f 3 They are the known nonlinear parts of the three-axis angular rate change channel, D φ ,D θ ,D ψ are the lumped interferences of the three channels respectively; the attitude angle tracking error system is written as:

[0050]

[0051] Among them They are respectively the second-order derivative of the roll angle with respect to time, the second-order derivative of the pitch angle with respect to time, and the second-order derivative of the yaw angle tracking error with respect to time;

[0052] Considering the constraint on the attitude angle change rate, the design idea is introduced by taking the roll angle as an example. The roll angle change rate constraint is set as Where m φ2 ,m φ1is a known constant; for the constant value command, the roll angle change constraint is That is, the attitude angle change rate constraint is transformed into the attitude angle tracking error change rate constraint; for the case of tracking the constant attitude angle command, combined with the lumped disturbance estimator of the observer, the attitude loop composite anti-interference controller is designed as shown below:

[0053]

[0054] where e φ ,e θ ,e ψ are the rolling angle, pitch angle and yaw angle tracking errors respectively, e φ ,e θ ,e ψ The first derivative with respect to time is are the estimates of the three-axis lumped interference given by the observer; m φ1 ,m θ1 ,m ψ1 is a positive constant equal to the upper bound of the constraint, m φ2 ,m θ2 ,m ψ2 is a negative constant equal to the lower bound of the constraint; k φ1 ,k θ1 ,k ψ1 ,k φ2 ,k θ2 ,k ψ2 ,l φ ,l θ ,l ψ is the control parameter of the pole configuration process. The pole configuration result needs to be that the real part of its poles are all located on the left half of the imaginary axis, so that the tracking system is asymptotically stable. When adjusting the parameters of the actual system, the corresponding poles are configured considering the expected dynamic characteristics of the system and the corresponding normal number is selected.

[0055] According to formula (7) and (9), the required torque is obtained:

[0056]

[0057] Step 4 includes:

[0058] According to equation (10) and equation (3), the longitudinal period and transverse period corresponding to the required moment are obtained:

[0059]

[0060] According to the main rotor collective pitch δ col Combining formula (10) and formula (4), the tail rotor pitch corresponding to the required torque is obtained:

[0061] δped =(N n -N r R col δ col ) / N ped (12)

[0062] Finally, the three actual control variables of the unmanned helicopter attitude loop are obtained: tail rotor pitch δ ped , longitudinal periodic pitch δ lon , lateral periodic variation δ lat .

[0063] The present invention also provides a storage medium storing a computer program or instruction. When the computer program or instruction is executed, the active anti-disturbance fault-tolerant control method for an unmanned helicopter considering constraints is implemented.

[0064] The present invention has the following beneficial effects: 1. The attitude angle change rate constraint is fully considered when designing the controller, and soft processing of the constraint is achieved while giving full play to the control capability; 2. The system has the ability of active anti-interference and fault tolerance through estimation and feedforward compensation of the lumped interference; 3. The influence of faults and interference is fully considered in the modeling process, and a disturbed attitude rudder loop model containing actuator faults is established in combination with the servo characteristics; 4. The present invention improves the tracking accuracy of the attitude command of the unmanned helicopter in the event of an actuator failure, and its control concept can also be used in a variety of similar nonlinear systems, and has broad engineering application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the above and other advantages of the present invention will become more clear.

[0066] Figure 1 This is a structural block diagram of the active anti-disturbance fault-tolerant control of an unmanned helicopter taking into account constraints adopted by the present invention.

[0067] Figure 2 The attitude angle tracking response curves of the unmanned helicopter under the action of the baseline dynamic inverse controller (Nonlinear Dynamic Inverse Controller, NDIC), the composite dynamic inverse controller (Composite Nonlinear Dynamic Inverse Controller, CNDIC) and the proposed constrained composite dynamic gain controller (Composite Dynamic Gain Controller, CDGC).

[0068] Figure 3The present invention provides attitude change rate response curves of an unmanned helicopter under the action of a reference dynamic inverse controller (NDIC), a composite dynamic inverse controller (CNDIC) and a composite dynamic gain controller (CDGC) according to the method of the present invention.

[0069] Figure 4 The control input response curves of the attitude loop of an unmanned helicopter under the action of a reference dynamic inverse controller (NDIC), a composite dynamic inverse controller (CNDIC) and a composite dynamic gain controller (CDGC) according to the method of the present invention are shown. DETAILED DESCRIPTION

[0070] Example

[0071] This embodiment first uses the Simulink toolbox in the simulation software MATLAB R2018b to build an unmanned helicopter attitude loop model considering actuator failures, and then conducts simulations and experiments. Figure 1 As shown, the present invention provides an active anti-disturbance fault-tolerant control method for an unmanned helicopter considering constraints, and the specific steps include:

[0072] Step 1, establishing a disturbed nonlinear dynamic model of the attitude loop of an unmanned helicopter including actuator failure;

[0073] Step 2: Based on the extended state observer technique, an interference observer is designed to estimate the aggregate interference.

[0074] Step 3, designing a composite anti-interference controller for the attitude loop by combining the interference estimation information and the dynamic gain constraint processing control method;

[0075] Step 4, based on the required torque, the actual control variables of the helicopter, cyclic pitch and tail rotor pitch, are obtained by inverse solution.

[0076] Step 1 is as follows:

[0077] Establish a dynamic model of the attitude loop subject to multiple sources of interference:

[0078]

[0079] Where Ω=[p,q,r] T is the angular velocity of rotation around the body axis, p, q, r are the roll, pitch, and yaw angular rates respectively. J = diag(I x ,I y ,I z ) is the moment of inertia matrix, Γ=[L,M,N] T is the three-axis moment, L, M, N are rolling moment, pitch moment and yaw moment respectively, D A is a vector with three rows and one column, which represents the multi-source interference of the attitude loop. T(Θ) is the transformation matrix, which is specifically expressed as:

[0080]

[0081] Among them, sinφ=sφ, cosφ=cφ, tanθ=tθ, sinθ=sθ, cosθ=cθ.

[0082] Model the actuator including the effects of faults:

[0083] The relationship between rolling torque L, pitching moment M and periodic moment:

[0084]

[0085] In the above formula, L a ,L b ,M a ,M b ,A lon ,A lat ,B lon ,B lat are system parameters, ε, τ are time constants, are the failure coefficients of the longitudinal period distance and the transverse period distance, respectively. a =55.86,L b =708.02, M a =345.19,M b =-11.45,A lat =-5.57,A lon =-14.68,B lat =-19.46,B lon =-2.56.

[0086] For the convenience of writing, the following definitions are introduced:

[0087]

[0088] Then equation (2) can be rewritten as:

[0089]

[0090] Among them, L n ,M n is the nominal rolling torque and pitching moment when the cyclic distance is fault-free, L f ,M f Roll torque and pitch moment caused by cyclic fault:

[0091]

[0092] The relationship between yaw moment, tail rotor pitch and total pitch:

[0093]

[0094] N r ,N col ,N ped is a system parameter. In this embodiment, N r =-23.9,N col =8.89,N ped =113.65.δ ped is the tail rotor pitch, is the tail rotor pitch failure coefficient, N n is the nominal value of the yaw moment without considering the fault, N f Yaw moment caused by tail rotor failure:

[0095] N n =N r r+N col δ col +N ped δ ped ,

[0096] Step 2 specifically includes:

[0097] Based on the extended state observer technique, an interference observer is designed to estimate the aggregate interference.

[0098] Define attitude angle tracking error:

[0099]

[0100] According to equation (2), the tracking error dynamics of the attitude system is obtained:

[0101]

[0102] Where Γ n The nominal value of the torque around the fuselage axis when there is no fault in the cyclic pitch, collective pitch and tail pitch:

[0103] Γ n =[L n M n N n ] T

[0104] is the matrix after the derivative of each element of the T(Θ) matrix with respect to t, as follows:

[0105]

[0106] D in the formula LA ∈R 3×1 is the aggregate disturbance, including the disturbance, fault and the second-order derivative information of the attitude angle command:

[0107] Γ f=[L f M f N f ] T ,

[0108] For the tracking error system (5), an extended state observer is designed to estimate the aggregate interference:

[0109]

[0110] Where the observer gain matrix For R 3×3 In this embodiment, is an estimate of the aggregate interference.

[0111] Step 3 specifically includes:

[0112] To facilitate subsequent design, the following definition is introduced for the tracking error system (5):

[0113]

[0114] Then the attitude angle tracking error system is written as:

[0115]

[0116] Considering the attitude angle change rate constraint, the design idea is introduced by taking the roll angle as an example. Assume that the roll angle change rate constraint is Where m φ2 ,m φ1 is a known constant. In view of the situation of tracking the constant attitude angle command, the composite controller of the attitude system is designed by combining the lumped disturbance estimator of the observer:

[0117]

[0118] In the above formula, m φ1 ,m θ1 ,m ψ1 m φ2 ,m θ2 ,m ψ2 is a constant; control parameter k φ1 ,k φ2 ,k θ1 ,k θ2 ,k ψ1 ,k ψ2 ,l φ ,l θ ,l ψ All values ​​are normal numbers; for e φ ,e θ ,e ψIn this embodiment, m φ1 =m θ1 =45° / s,m ψ1 =30° / s,m φ2 =m θ2 =-45° / s,m ψ2 =-30° / s. In this embodiment, the control parameter is specifically set to k φ1 =k θ1 =k ψ1 =100,k φ2 =k θ2 =k ψ2 =20,l φ = l θ =5,l ψ =2.

[0119] According to definition (7) and controller (9), the required torque is obtained:

[0120]

[0121] Step 4 specifically includes:

[0122] According to equation (10) and equation (3), the longitudinal period and transverse period corresponding to the required moment are obtained:

[0123]

[0124] According to the main rotor collective pitch δ col Combined with equation (10) and equation (4), the tail rotor pitch corresponding to the required torque is obtained:

[0125] δ ped =(N n -N r R col δ col ) / N ped (12)

[0126] The three actual control quantities of the unmanned helicopter attitude loop are obtained: tail rotor pitch δ ped , longitudinal periodic pitch δ lon , lateral periodic variation δ lat .

[0127] Simulation verification:

[0128] In order to verify the anti-interference performance of the composite controller, the proposed method is simulated and verified based on MATLAB / Simulink. Under the condition of fully considering the actuator failure and external time-varying interference, the dynamic inverse controller (Nonlinear Dynamic Inverse Controller, NDIC), the composite dynamic inverse controller (Composite Nonlinear Dynamic Inverse Controller, CNDIC) and the composite dynamic gain controller (Composite Dynamic Gain Controller, CDGC) proposed in this invention are selected for comparison.

[0129] Initial state of the unmanned helicopter attitude loop:

[0130] φ(0)=0,θ(0)=0,ψ(0)=0

[0131] p(0)=0,q(0)=0,r(0)=0

[0132] Attitude angle command:

[0133] φ r =15°,θ r =15°,ψ r =10°

[0134] The actuator fault settings are as follows:

[0135] 0s <t≤8s:

[0136] 8s <t≤15s:

[0137] Torque channel interference settings:

[0138]

[0139] The attitude angle change rate constraint is set as:

[0140] m φ1 =m θ1 =45° / s,m ψ1 =30° / s; m φ2 =m θ2 =-45° / s,m ψ2 =-30° / s

[0141] Controller parameter setting:

[0142] k φ1 =k θ1 =k ψ1 =100; k φ2 =kθ2 =k ψ2 =20; l φ = l θ =5,l ψ =2

[0143] Simulation analysis: Figure 2 and Figure 3 The attitude angle response curve and attitude change rate response curve of the unmanned helicopter under the three methods of dynamic inverse controller NDIC, composite dynamic inverse controller CNDIC and composite dynamic gain controller CDGC with constraints proposed by the present invention. Figure 2 It can be seen that the posture tracking accuracy of the two composite control methods, CNDIC and CDGC, is significantly better than that of the NIDC control method, and the tracking accuracy of CNDIC is slightly higher than that of the proposed CDGC method. Figure 3 It can be seen that only the proposed method ensures that the attitude angle change rate is within the constraint range, while the attitude angle change rates of the other two methods exceed the constraints. Figure 4 These are the input response curves of the attitude loop control of the unmanned helicopter under the three control methods. It can be seen from the figure that the control quantities of the three control methods are at the same level.

[0144] In a specific implementation, the present application provides a computer storage medium and a corresponding data processing unit, wherein the computer storage medium can store a computer program, and when the computer program is executed by the data processing unit, the invention content of the active anti-disturbance fault-tolerant control method for an unmanned helicopter considering constraints provided by the present invention and some or all of the steps in each embodiment. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0145] Those skilled in the art can clearly understand that the technical solutions in the embodiments of the present invention can be implemented by means of computer programs and their corresponding general hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention can be essentially or partly contributed to the prior art in the form of a computer program, i.e., a software product, which can be stored in a storage medium and includes several instructions for enabling a device including a data processing unit (which can be a personal computer, a server, a single-chip microcomputer, a MUU or a network device, etc.) to execute the methods described in various embodiments of the present invention or certain parts of the embodiments.

[0146] The present invention provides an active anti-disturbance fault-tolerant control method for an unmanned helicopter considering constraints. There are many methods and ways to implement the technical solution. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the protection scope of the present invention. All components not specified in this embodiment can be implemented by existing technologies.

Claims

1. An active anti-disturbance fault-tolerant control method for an unmanned helicopter considering constraints, characterized in that: The following steps are involved: Step 1, establishing a disturbed nonlinear dynamic model of the attitude loop of an unmanned helicopter including actuator failure; Step 2: Based on the extended state observer technique, an interference observer is designed to estimate the aggregate interference. Step 3, designing a composite anti-interference controller for the attitude loop by combining the interference estimation information and the dynamic gain constraint processing control method; Step 4, based on the required torque, the actual control quantity of the helicopter, cyclic pitch and tail rotor pitch, is obtained by inverse solution; Step 1 contains the following steps: Step 1-1, establishing a posture loop dynamics model affected by multiple sources of interference; Step 1-2, establish an actuator model including the impact of the fault; In step 1-1, the attitude loop dynamics model affected by multi-source interference is as follows: where Θ = [φ,θ,ψ] T is the helicopter attitude angle, φ, θ, ψ are roll angle, pitch angle, and yaw angle respectively. is the column vector of the first-order derivative of the helicopter attitude angle with respect to time, They are the first-order derivative of the roll angle with respect to time, the first-order derivative of the pitch angle with respect to time, and the first-order derivative of the yaw angle with respect to time; Ω = [p, q, r] T is the angular velocity of rotation around the body axis, p, q, r are roll angle, pitch angle, and yaw rate respectively; J = diag (I x ,I y ,I z ) is the moment of inertia matrix, I x ,I y ,I z are the moments of inertia around the three axes of the fuselage; Γ=[L,M,N] T is the three-axis moment, L, M, N are rolling moment, pitch moment and yaw moment respectively, D A is the multi-source interference received by the attitude loop, which is a vector with three rows and one column; T(Θ) is the transformation matrix, and its specific expression is: Among them, sinφ=sφ, cosφ=cφ, tanθ=tθ, sinθ=sθ, cosθ=cθ; Steps 1-2 include: The relationship between the rolling torque L, the pitching moment M and the periodic moment is as follows: Where L a ,L b ,M a ,M b ,A lon ,A lat ,B lon ,B lat are system parameters, ε, τ are time constants, are the failure coefficients of the longitudinal period distance and the transverse period distance, respectively, δ lon ,δ lat are the longitudinal periodic distance and the transverse periodic distance respectively; Introduce the intermediate parameters A0, A1, Φ, Then equation (2) can be rewritten as: Among them, L n ,M n They are respectively the nominal rolling torque and pitching moment when the period distance is fault-free, L f ,M f They are respectively the rolling torque and pitching moment caused by cyclic pitch fault, as shown below: The relationship between the yaw moment, tail rotor pitch and total pitch is as follows: Where N r ,N col ,N ped is the system parameter, δ col is the main rotor pitch, is the main rotor pitch failure coefficient, δ ped is the tail rotor pitch, is the tail rotor pitch failure coefficient, N n is the nominal value of the yaw moment without considering the fault, N f The yaw moment caused by tail rotor failure is as follows: Step 2 includes: defining the attitude angle tracking error: Where e Θ is the attitude angle tracking error column vector, Θ is the attitude angle vector, Θ d =[φ d ,θ d ,ψ r ] T is a three-row and one-column attitude angle command vector, φ d ,θ d ,ψ r They are roll angle command, pitch angle command and yaw angle command respectively; According to equation (2), the tracking error dynamics of the attitude system is obtained: in For e Θ The column vector of the second-order derivatives of each element with respect to time, Γ n is the nominal value of the torque around the fuselage axis when there is no fault in the cyclic pitch, collective pitch and tail pitch, L n ,M n ,N n They are the nominal values ​​of rolling torque, pitching moment and yaw moment respectively: C n =[L n M n N n ] T is the matrix after the derivative of each element of the T(Θ) matrix with respect to t, as follows: Among them, Γ f is the torque variation caused by the total pitch, tail rotor pitch and periodic pitch change faults, specifically L f ,M f ,N f They are respectively the change of rolling moment, the change of pitch moment and the change of yaw moment; D is the column vector of the second-order derivative of the attitude angle command vector with respect to time. LA It is a vector with three rows and one column, which is the lumped disturbance, including the information of disturbance, fault and the second-order derivative of attitude angle command: For formula (5), the interference observer is designed to estimate the aggregate interference as follows: Where z1, z2, z3 are the state variables of the observer. is the first-order derivative of z1, z2, z3 with respect to time, and the observer gain matrix For R 3×3 The diagonal matrix of is the estimate of the aggregate interference; Step 3 includes: For formula (5), the following definition is introduced: Where τ1, τ2, τ3 are virtual rolling torque, virtual pitch moment and virtual yaw moment respectively, f1, f2, f3 are the known nonlinear parts of the three-axis angular rate change channel respectively, and D φ ,D θ ,D ψ are the lumped interferences of the three channels respectively; the attitude angle tracking error system is written as: Among them They are respectively the second-order derivative of the roll angle with respect to time, the second-order derivative of the pitch angle with respect to time, and the second-order derivative of the yaw angle tracking error with respect to time; Set the roll angle change rate constraint to Where m φ2 ,m φ1 is a known constant; for the constant value command, the roll angle change constraint is That is, the attitude angle change rate constraint is transformed into the attitude angle tracking error change rate constraint; for the case of tracking the constant attitude angle command, combined with the lumped disturbance estimator of the observer, the attitude loop composite anti-interference controller is designed as shown below: where e φ ,e θ ,e ψ are the rolling angle, pitch angle and yaw angle tracking errors respectively, e φ ,e θ ,e ψ The first derivative with respect to time is are the estimates of the three-axis lumped interference given by the observer; m φ1 ,m θ1 ,m ψ1 is a positive constant equal to the upper bound of the constraint, m φ2 ,m θ2 ,m ψ2 is a negative constant equal to the lower bound of the constraint; k φ1 ,k θ1 ,k ψ1 ,k φ2 ,k θ2 ,k ψ2 ,l φ ,l θ ,l ψ Control parameters of the process for configuring poles; According to formula (7) and (9), the required torque is obtained:

2. The method according to claim 1, characterized in that Step 4 includes: According to equation (10) and equation (3), the longitudinal period and transverse period corresponding to the required moment are obtained: According to the main rotor collective pitch δ col Combining formula (10) and formula (4), the tail rotor pitch corresponding to the required torque is obtained: d ped =(N n -N r rN col d col ) / N ped (12) Finally, the three actual control variables of the unmanned helicopter attitude loop are obtained: tail rotor pitch δ ped , longitudinal periodic pitch δ lon , lateral periodic variation δ lat .

3. A storage medium, characterized in that: A computer program or instruction is stored, and when the computer program or instruction is executed, the method according to any one of claims 1 to 2 is implemented.

Images