Unmanned aerial vehicle preset performance control method and system considering time-varying position constraint

Abstract

The invention belongs to the technical field of unmanned aerial vehicle trajectory tracking control, and particularly discloses an unmanned aerial vehicle preset performance control method and system considering time-varying position constraints. According to the method, under the condition that the unmanned aerial vehicle has time-varying position constraint, model uncertainty, actuator fault and wind disturbance, a nonlinear mapping method is firstly adopted to convert a system with position constraint into an unconstrained system, then a preset performance control method and a self-adaptive backstepping technology are adopted to carry out controller design on the converted system, and therefore the control precision of the unmanned aerial vehicle is improved. The system error is kept in a preset boundary function range; the invention also designs a fuzzy logic system (FLS) and a nonlinear disturbance observer to estimate the uncertainty and wind disturbance of the model, thereby providing a trajectory tracking control scheme combining model mapping and preset performance. Under the control of the method, the quad-rotor unmanned aerial vehicle has good robustness, and even if an actuator breaks down, the tracking error can still be kept within a preset interval.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicle trajectory tracking control, and in particular relates to a method and system for controlling the preset performance of an unmanned aerial vehicle taking into account time-varying position constraints. Background Art

[0002] In recent decades, drones have increasingly appeared in the public eye, such as rescue, agriculture, photography, etc. As the most popular type of wing-type drones, quadrotor drones (Unmanned Aerial Vehicle, UAV) have many unique advantages such as strong flight stability and adaptability, vertical take-off and landing, and easy production. Quadrotor drones are not only used in wide areas, but also in constrained narrow areas such as communities and pipelines, which puts higher requirements on drone systems. The four-input and six-output quadrotor drone system is considered to be an under-actuated and highly coupled system, which makes control more difficult. Therefore, how to design a constrained drone controller with high adaptability and high efficiency becomes a challenging task.

[0003] Experts at home and abroad have conducted in-depth research on the design of complex nonlinear systems and constrained systems. Model predictive control methods are considered to be a powerful means to solve constrained control problems. This method is essentially an optimization strategy, but this strategy requires a lot of calculations and is not easy to implement. Barrier Lyapunov Function (BLF) technology is usually used to deal with state constraint problems. In previous studies, experts and scholars have proposed a non-singular fast terminal sliding mode controller based on BLF to ensure the finite time convergence of the Blended Wing Body (BWB) aircraft system. In addition, an adaptive model-free controller based on a time-varying BLF of a manipulator has also been proposed. The BLF method is used in all of the above studies. However, the BLF method requires some necessary assumptions, such as the virtual control input needs to meet the constraints, but this is difficult in practice. Unlike the use of the BLF method for constrained systems, more and more scholars have begun to study methods such as nonlinear mapping functions and coordinate transformations. At present, scholars have proposed the use of nonlinear mapping methods to design controllers for constrained robot systems, and some literature has introduced a new mapping function to replace the BLF method, which also eliminates the assumption that the virtual control input meets the constraints.

[0004] In engineering practice, with the continuous improvement of control performance requirements, Prescribed Performance Control (PPC) came into being and gradually became popular. Among them, for UAVs with disturbances, experts have proposed an adaptive prescribed performance controller that can ensure that the system error converges quickly within the prescribed range. The PPC method can also solve the tracking problem of vehicles with actuator failures and ensure that the tracking error converges to the prescribed area. In addition, the use of preset performance technology to design an observer to solve the problem of frequent actuator failures of UAVs can make the system more robust and adaptive. However, in the face of time-varying constraints on the system state, the design of the preset performance function becomes complicated.

[0005] As we all know, quadrotor drones are extremely susceptible to actuator failures, which can lead to control deterioration or even failure. Therefore, the problem of actuator failure has attracted widespread attention. In previous studies, a new adaptive control method based on a state estimator was proposed to solve the actuator failure problem of uncertain delayed semi-Markov jump systems, but the time-varying position constraint problem was not considered. In addition, scholars studied a fault-tolerant control for handling drone thrust system failures. Specifically, when the control output loss is serious, the direction control can be abandoned to maintain attitude controllability. However, this method is not very applicable in strict position constraints. In addition, experts have also designed a high-gain observer that can effectively compensate for the negative impact of quadrotor drone actuator failures, but this method has not been well applied to position constraint problems. Summary of the invention

[0006] The purpose of the present invention is to propose a preset performance control method for an unmanned aerial vehicle (UAV) considering time-varying position constraints. The method targets a complex system with time-varying position constraints, model uncertainties, external wind disturbances and actuator failures. The method transforms the constraint model into a new unconstrained model using a nonlinear mapping function, and designs a preset performance controller for the UAV using preset performance control and adaptive backstepping control, so as to enable the UAV to stably follow the expected signal without violating the constraints, and to achieve higher control accuracy and robustness.

[0007] In order to achieve the above object, the present invention adopts the following technical scheme:

[0008] The preset performance control method of the UAV considering the time-varying position constraint includes the following steps:

[0009] Step 1. Establish a composite model of the UAV that takes into account time-varying position constraints, model uncertainty, and external wind disturbances, and use nonlinear mapping relationships to map the composite model and convert it into an unconstrained model; use a preset performance function to limit the tracking error of the unconstrained system, use an error transformation function to process the tracking error, and use it for the design of the subsequent preset performance controller of the UAV;

[0010] Step 2. Based on the unconstrained system in step 1, the model uncertainty is estimated using an adaptive fuzzy system, and the external wind disturbance and actuator failure are estimated and compensated using a disturbance observer. The preset performance method and backstepping method are used to design the UAV preset performance controller.

[0011] Step 3. Use the preset controller of the UAV designed in step 2 to realize trajectory tracking control of the unconstrained system obtained after nonlinear mapping, and according to the nonlinear mapping relationship, the expected signal of the original constrained system to stably follow without violating the constraints can be obtained.

[0012] In addition, based on the above, the present invention also proposes a method for controlling the preset performance of a UAV considering time-varying position constraints, which adopts the following technical solutions:

[0013] The preset performance control system of the quadrotor drone considering time-varying position constraints includes the following modules:

[0014] The model building module is used to establish a composite model of the UAV that takes into account time-varying position constraints, model uncertainties, and external wind disturbances, and to map the composite model using a nonlinear mapping relationship to convert it into an unconstrained model; to limit the tracking error of the unconstrained system using a preset performance function, and to process the tracking error using an error transformation function, and to use it in the design of a subsequent preset performance controller for the UAV;

[0015] The controller design module is used to estimate model uncertainty based on an unconstrained system using an adaptive fuzzy system, estimate and compensate for external wind disturbances and actuator failures using a disturbance observer, and design a UAV preset performance controller using the preset performance method and backstepping method;

[0016] And an actuator fault control module is used to use the UAV preset controller to realize trajectory tracking control of the unconstrained system obtained after nonlinear mapping, and according to the nonlinear mapping relationship, the expected signal for the original constrained system to stably follow without violating the constraint conditions can be obtained.

[0017] In addition, based on the above-mentioned method for controlling the preset performance of a UAV taking into account time-varying position constraints, the present invention also proposes a computer device; the computer device includes a memory and one or more processors, and the memory stores executable code. When the processor executes the executable code, the steps of the method for controlling the preset performance of a UAV taking into account time-varying position constraints as described above are implemented.

[0018] In addition, based on the above-mentioned method for controlling the preset performance of a UAV taking into account time-varying position constraints, the present invention also proposes a computer-readable storage medium; a program is stored on the computer-readable storage medium, and when the program is executed by the processor, the steps of the method for controlling the preset performance of a UAV taking into account time-varying position constraints as described above are implemented.

[0019] The present invention has the following advantages:

[0020] As described above, the present invention relates to a method for controlling the preset performance of a drone considering time-varying position constraints, and adopts a method combining nonlinear mapping and preset performance control for a quadrotor drone under constraint conditions. The method uses a nonlinear mapping function to cleverly convert a model of a drone with position constraints into an unconstrained model. Compared with the BLF method, the method avoids the problems of complex BLF design and narrow initial output range. In addition, the preset performance control method of a drone considering time-varying position constraints of the present invention does not require the assumption that the virtual control input satisfies the constraint boundary. For drones with model uncertainty and external interference, a fault-tolerant adaptive preset performance controller that can adapt to actuator failures and position constraints is designed by combining PPC with backstepping control. Through the designed controller, the drone trajectory tracking error can be effectively kept within the preset performance boundary range. The disturbance observer and FLS can effectively compensate for the adverse effects of external interference and model uncertainty, respectively, so that the quadrotor drone under the control of the method of the present invention has good robustness and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flow chart of a method for controlling the preset performance of a drone taking into account time-varying position constraints in an embodiment of the present invention;

[0022] Figure 2 A schematic diagram of a nonlinear mapping function in an embodiment of the present invention;

[0023] Figure 3 A system block diagram of a method for controlling the preset performance of a drone taking into account time-varying position constraints in an embodiment of the present invention;

[0024] Figure 4 A schematic diagram of actuator fault parameters in an embodiment of the present invention;

[0025] Figure 5This is a schematic diagram of output response in an embodiment of the present invention;

[0026] Figure 6 This is a 3D actual motion trajectory tracking diagram of the drone in an embodiment of the present invention;

[0027] Figure 7 It is a tracking diagram of the actual position state trajectory in an embodiment of the present invention;

[0028] Figure 8 Schematic diagram of actual position state error change in an embodiment of the present invention;

[0029] Fig. 9 It is a schematic diagram of tracking the conversion state trajectory of the position state in an embodiment of the present invention;

[0030] Fig.10 Schematic diagram of the change of the conversion state error of the position state in an embodiment of the present invention;

[0031] Fig.11 This is a schematic diagram of actual posture state trajectory tracking in an embodiment of the present invention;

[0032] Fig.12 Schematic diagram of the change of posture state error in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0034] Example 1

[0035] In view of the deficiencies in the research of the above-mentioned background technology, the present invention aims at the control problem of quadrotor UAV under time-varying position constraints. It uses a nonlinear mapping method to avoid the "feasibility condition" assumption of the BLF method, uses a preset performance control method to design a controller, and takes into account the influence of actuator failure and external wind disturbance, effectively ensuring the stable flight of the quadrotor UAV under time-varying position constraints.

[0036] Aiming at the actuator failure system of a quadrotor drone with time-varying position constraints, model uncertainty and external disturbances, this embodiment proposes a drone preset performance control method considering time-varying position constraints. In order to achieve high-precision trajectory tracking control, this method proposes a scheme combining model nonlinear mapping and preset performance control, uses an adaptive fuzzy system to estimate the unknown internal dynamics of the model, and designs a nonlinear disturbance observer to estimate external disturbances and actuator failures, making it more suitable for practical applications. Finally, through experimental analysis, the control performance of the proposed method for a quadrotor drone with time-varying position constraints was verified.

[0037] Based on the analysis and research of the existing technology, the present invention fully integrates a variety of theoretical technologies such as backstepping control, fuzzy logic system, Lyapunov stability theory, etc., and focuses on designing a preset performance controller for the quadrotor UAV to address the problems of time-varying position constraints, actuator failures, model uncertainty and external wind disturbances. The nonlinear mapping method is combined with the PPC method to effectively solve the time-varying constraint problem of the quadrotor UAV.

[0038] The control method of the present invention is aimed at a quad-rotor unmanned aerial vehicle. The control goal is to use nonlinear function mapping and preset performance control methods to achieve stable following of expected signals by the unmanned aerial vehicle under the condition of spatial position constraints. Considering the spatial position constraints, namely the position state constraints, not only the tracking errors of all states are constrained, but also the states of the position subsystems are subject to more normal time-varying constraints. The present invention uses a nonlinear mapping method and coordinate transformation to deal with the position constraint problem, thereby avoiding the "feasibility condition" assumption in the BLF method. In addition, the method of the present invention uses a nonlinear function to convert the position state constraint into an unconstrained model and then performs preset performance control, and uses a disturbance observer to observe and compensate for the adverse effects of external disturbances, which is more universal and practical.

[0039] like Figure 1 As shown in FIG. 1 , the preset performance control method of the UAV considering the time-varying position constraint specifically includes the following steps:

[0040] Step 1. Establish a composite model of the quadrotor UAV that considers time-varying position constraints, model uncertainties, and external wind disturbances, and use nonlinear mapping relationships to map the composite model and convert it into an unconstrained model; use preset performance functions to limit the tracking error of the unconstrained system, use error transformation functions to process the tracking error, and use it for the design of subsequent UAV preset performance controllers.

[0041] Assume that the position vector P and the Euler angle vector Θ are represented as P = [x, y, z] T ∈R 3 and Θ = [φ, θ, ψ] T ∈R 3 , where (x, y, z) is the position of the UAV in the inertial coordinate system, and φ, θ, and ψ represent the roll angle, pitch angle, and yaw angle of the UAV, respectively.

[0042] To facilitate the subsequent processing of the position subsystem, let [P1(t), P2(t), P3(t)] T =[x,y,z] T , where P1(t) represents x, P2(t) represents y, and P3(t) represents z. The definition of P here is for the convenience of using subscripts to represent x, y, and z in the subsequent design of the preset performance controller of the drone.

[0043] Establish the UAV position subsystem model as shown in formula (1a) and the UAV attitude subsystem model as shown in formula (1b):

[0044]

[0045] Where A=[A1,A2,A3] T , A1=(cosφsinθcosψ+sinφsinψ) / m, A2=(cosφsinθsinψ-sinφcosψ) / m, A3=cosφsinθ / m, m represents the total mass of the drone; H a represents the unknown nonlinear composite term acting on the position subsystem, H a =[H1,H2,H3] T ∈R 3 , where H1, H2, and H3 represent the unknown nonlinear composite terms acting on the x-axis, y-axis, and z-axis directions of the position subsystem, respectively; d a (t) represents the external wind disturbance to the position subsystem, d a (t) = [d1, d2, d3] T ∈R 3 , where d1, d2, and d3 represent the external wind disturbances experienced by the position subsystem in the x-axis, y-axis, and z-axis directions, respectively.

[0046] J (·) represents the moment of inertia, J x , J y , J z They represent the UAV's moment of inertia about the x-axis, y-axis, and z-axis respectively, l is the distance from the UAV propeller to the center of mass; u Θ =[u o,2 ,u o,3 ,u o,4 ] T ∈R 3 ,u o,1 、u o,2 、u o,3 、u o,4 They represent the control inputs generated by the four motor rotors of the drone; H b represents the unknown nonlinear composite term acting on the attitude subsystem, H b =[H4,H5,H6] T ∈R 3 , where H4, H5, and H6 represent the unknown nonlinear composite terms acting on the roll angle, pitch angle, and yaw angle of the attitude subsystem, respectively; d b (t) represents the external wind disturbance to the attitude subsystem, d b (t) = [d4, d5, d6] T∈R 3 , where d4, d5, and d6 represent the external wind disturbances on the roll angle, pitch angle, and yaw angle of the attitude subsystem, respectively.

[0047] The UAV position subsystem and attitude subsystem models have the influence of asymmetric time-varying position constraints as shown in formula (2):

[0048]

[0049] Among them, i1=1,2,3, Represents a collection, is a known positive function.

[0050] H (·) =f (·) +△f (·) ;

[0051] Among them, H (·) Represents the unknown nonlinear composite term acting on the position subsystem and attitude subsystem.

[0052] f (·) represents the air resistance term, where f1, f2, f3, f4, f5, and f6 represent the air resistance in the x-axis, y-axis, z-axis, roll angle, pitch angle, and yaw angle directions, respectively. g is the acceleration due to gravity; G (·) is the air resistance coefficient, where G x , G y , G z , G φ , G θ , G ψ They represent the air resistance coefficient of the drone in the x-axis, y-axis, z-axis, roll angle, pitch angle and yaw angle directions respectively.

[0053] △f (·) represents the model uncertainty, [△f1,△f2,△f3] T ∈R 3 and [△f4,△f5,△f6] T ∈R 3 They represent the model uncertainties of the position subsystem and attitude subsystem respectively, where △f1, △f2, and △f3 represent the model uncertainties of the position subsystem in the directions of the x-axis, y-axis, and z-axis respectively, and △f4, △f5, and △f6 represent the model uncertainties of the attitude subsystem in the directions of the roll angle, pitch angle, and yaw angle respectively.

[0054] u o,1 、u o,2 、u o,3 、u o,4 The relationship is:

[0055]

[0056] Among them, l1 is the lift coefficient, l2 is the rotor drag coefficient, and w (·) is the motor rotor speed, w1, w2, w3, and w4 represent the four motor rotor speeds of the drone respectively.

[0057] In actual engineering, in order to avoid singularity problems, the roll angle and pitch angle should satisfy and The external wind disturbances d1, d2, d3, d4, d5, and d6 are unknown and bounded; the time-varying position constraint boundaries do not cross, that is, the upper constraint bound is always greater than the lower constraint bound.

[0058] Let the expected trajectory [Υ1, Υ2, γ3] T =[x d ,y d ,z d ] T , where x d ,y d 、z d represents the expected position trajectory of the drone in the inertial coordinate system. and Make the expected trajectory and its derivative satisfies and And for the desired yaw angle ψ d Just meet Among them, i1=1,2,3, j1=1,2. In addition, the system states x, y, z, θ, ψ, φ can be used to design the preset performance controller of the quadrotor drone.

[0059] When the four actuators of the drone may be affected by failure during operation, the actual input With expected input There are differences, consider the following relationship:

[0060]

[0061] Among them, i2=1,2,3,4, and is a continuous time-varying function, and represent the partial fault factor and bias fault factor respectively, and and are all unknown and bounded continuous functions.

[0062] For a quadcopter drone system with external constraints, it is challenging to directly design a controller. To this end, this paper proposes a method for preprocessing the constraint model with a nonlinear mapping function to transform the constraint model into an unconstrained model. In a nonlinear system, assuming that the constraint state is s, the mapped new state is set to ζ, and the constraint range is -k c1 <s<k c2 , where k c1 , k c2 are all positive numbers. Consider the following smooth nonlinear mapping expression N(k c1 ,k c2 ,s):s→ζ, where N(·) is a nonlinear mapping function and the expression of ζ is shown in formula (5):

[0063]

[0064] Depend on Figure 2 It can be seen that when s approaches the constraint boundary, ζ→∞. The range of values ​​of the mapped new state ζ is unconstrained, that is, ζ∈R.

[0065] Defining system status And suppose [x1,x3,x5] T The actual position state [x,y,z] T The conversion state obtained by the nonlinear mapping function N(·), that is, x1=N(F 11 ,F 12 ,x),x3=N(F 21 ,F 22 ,y),x5=N(F 31 ,F 32 ,z). Obviously, through the transformation of the nonlinear mapping function, we get [x1,x3,x5] T ∈R 3 .

[0066] In order to obtain the transformed unconstrained model, we obtain The derivative of is:

[0067]

[0068] Among them, i1=1,2,3, and The expression is as follows:

[0069]

[0070] Obviously, in the previously defined set Medium, μ (·)is positive and computable for control design. Combining the UAV dynamics model shown in formula (1a) and formula (1b), the time-varying position constraint shown in formula (2), the actuator failure shown in formula (4), and the nonlinear mapping function N(·), we can obtain the unconstrained UAV position subsystem model S1 shown in formula (9a) and the unconstrained UAV attitude subsystem model S2 shown in formula (9b):

[0071]

[0072] Among them, [x1,x3,x5] T is the unconstrained state after conversion, i.e., the conversion state; H (·) Given in formula (1); u x 、u y and u z Represent the virtual control inputs of the x-subsystem, y-subsystem, and z-subsystem, respectively, and u x =u1(cosφsinθcosψ+sinφsinψ),u y =u1(cosφsinθsinψ-sinφcosψ),

[0073] u z =u1cosφcosθ.

[0074] D (·) Denotes the nonlinear term containing disturbance, where D1, D2, D3, D4, D5, and D6 represent the nonlinear terms containing disturbance in the x-axis, y-axis, z-axis, roll angle, pitch angle, and yaw angle directions, respectively. D1 = d1 + A1o1 - A1Γ1u1, D2 = d2 + A2o1 - A2Γ1u1, D3 = d3 + A3o1 - A3Γ1u1, D4 = d4 + K2o2 - K2Γ2u2, D5 = d5 + K3o3 - K3Γ3u3, and D6 = d6 + K4o4 - K4Γ4u4. Note that D (·) Depends on external wind disturbance, system state, fault parameters and expected input. The system state depends on the control input, and the control input that drives the drone depends on the speed of the four rotors. In actual engineering, it is assumed that the speed derivatives of the four rotors are bounded and there is a positive constant D ( * ·) Make

[0075] The position subsystem shown in formula (9a) is a nonlinear mapping subsystem, and the position controller is designed based on the unconstrained state. Therefore, the expected position signal also needs to be mapped as follows:

[0076]

[0077] Where i1 = 1, 2, 3, [Y1, Y2, Y3] T represents the expected position signal after mapping processing;

[0078] Assume that the expected gesture signal is [Y4, Y5, Y6] T =[φ d ,θ d ,ψ d ] T It is used for the design of subsequent UAV preset performance controller, where φ d ,θ d , d They represent the desired roll angle, pitch angle, and yaw angle of the drone, respectively. At the same time, position and rotation tracking errors are defined for:

[0079]

[0080] Among them, i3=1,3,5,7,9,11.

[0081] The above nonlinear mapping method has some important implications. When i1 = 1, 2, 3, for any Where t represents time, represents the initial position of x, y, and z. If for any t, are all bounded, then it can be guaranteed that for any t, Thus, the position constraint problem is reduced to ensuring The boundedness problem.

[0082] The preset performance methods are as follows:

[0083] -ρ(t) <z(t)<ρ(t)(12)

[0084] Wherein, z(t) is the tracking error signal of the unconstrained system shown in formulas (9a) and (9b), and ρ(t) is the preset performance function, which is selected as follows:

[0085] ρ(t)=(ρ0-ρ ∞ ) -ht +ρ ∞ (13)

[0086] Among them, ρ0, ρ ∞ , h is a positive constant, and satisfies ρ0>ρ ∞ In order to achieve the preset performance control of the tracking error, the error transformation function G(ε) is introduced:

[0087] z(t)=ρ(t)G(ε)(14)

[0088]

[0089] Among them, G(ε) is a differentiable function with positive derivative, ε is the new conversion error, and the expression of ε is:

[0090]

[0091] Taking the derivative of the new conversion error ε, we get:

[0092]

[0093] in and

[0094] The purpose of conversion error processing is to ensure that the tracking error z(t) is within the predetermined performance requirements and to prove that the conversion error ε(t) is bounded.

[0095] Step 2. Based on the unconstrained system in step 1, the model uncertainty is estimated using an adaptive fuzzy system, and the external wind disturbance and actuator failure are estimated and compensated using a disturbance observer. The preset performance method and backstepping method are used to design the UAV preset performance controller.

[0096] Step 2.1. Design a fuzzy observer consisting of an adaptive fuzzy system and a disturbance observer.

[0097] Since the nonlinear terms and model uncertainties in the models shown in formula (9a) and formula (9b) are unknown, the controller cannot be designed. To solve the above problem, the composite term Perform fuzzy approximation, where i3 = 1, 3, 5, 7, 9, 11, and obtain:

[0098]

[0099] in, represents an adaptive fuzzy system, represents the weight, represents the Gaussian basis function, Represents the approximation error. In the process of adaptive fuzzy system approximating the uncertainty of the model, the parameters are continuously updated, and the optimal weights Indicates the parameters for which the approximation is optimal.

[0100] Using adaptive fuzzy systems to approximate nonlinear terms, It cannot be directly used for the design of UAV preset performance controller, but it can be used for the design of UAV preset performance controller after approximating its value through fuzzy system. This process reflects the uncertainty of the estimation model.

[0101] In order to reduce the negative impact of the disturbance compound term, the present invention designs a fuzzy observer to approximate the disturbance compound term and defines Let be a bounded composite term containing perturbations. In order to conveniently represent the parameters in the fuzzy observer design process, we define j o =x,y,z,2,3,4 and K o =[K1,K1,K1,K2,K3,K4] T They correspond to i3=1,3,5,7,9,11 respectively, that is, when i3=1, j o =x,K o =K1; when i3=3, j o =y,K o =K1; when i3=5, j o = z, K o =K1; i3 = 7, j o =2, K o =K2; i3 = 9, j o =3, K o =K3; when i3=11, j o =4, K o =K4.

[0102] Design the fuzzy observer as shown in formula (18):

[0103]

[0104] in, yes The estimate, and are the design parameters and observer states, respectively, and yes estimates;

[0105] right The derivative is:

[0106]

[0107] in, express The estimation error of

[0108] set up for The estimation error of is obtained according to formula (19):

[0109]

[0110] Applying Young's inequality to the last term in equation (20) yields:

[0111]

[0112] in, is an unknown positive constant and satisfies There is an unknown positive constant satisfy

[0113]

[0114] Substituting formula (21) into formula (20), we obtain:

[0115]

[0116] in, Step 2.2. Design the preset performance controller of the quadrotor UAV using the preset performance method and backstepping method.

[0117] When i5=1,3,5 corresponds to j2=x,y,z, the position subsystem considered in formula (9a) is as follows:

[0118]

[0119] Define the state transition as:

[0120]

[0121] in, is the new state quantity; Tracking error The error conversion, and is the virtual control input. The z(t) in the above text is a general description of the preset performance method. This is the specific application of the method of the present invention in the process of UAV control design. Since it is combined with the backstepping design method, subscripts are required to distinguish each step.

[0122] Select the Lyapunov function as follows:

[0123]

[0124] Through formula (15) and formula (24), we can get The derivative of is:

[0125]

[0126] in, and The nonlinear expression term generated during the preset performance method is represented by r and v in the above text. and This is the specific application of the method of the present invention in the process of UAV control design. Since it is combined with the backstepping design method, subscripts are required to distinguish each step. and

[0127] Design the virtual control input as:

[0128]

[0129] in is the design parameter. Substituting formula (27) into formula (26), we get:

[0130]

[0131] Select the Lyapunov function as follows:

[0132]

[0133] right The derivative is:

[0134]

[0135] in,

[0136] Select the Lyapunov function as follows:

[0137]

[0138] right The derivative is:

[0139]

[0140] in, represents the design parameters; and i (·) It is given in formula (22), that is is an unknown positive constant and satisfies There is an unknown positive constant satisfy

[0141] When i5 = 1, 3, 5, the position subsystem controller is as shown in formula (33): And the fuzzy adaptive law shown in formula (44):

[0142]

[0143] in and is the design parameter, and Substituting formula (33) and formula (34) into formula (32), we obtain:

[0144]

[0145] According to Young's inequality, the following inequality holds:

[0146]

[0147] Substituting formula (36) and formula (37) into formula (35), we obtain:

[0148]

[0149] in,

[0150] Similar to the position subsystem controller design, when i6 = 7, 9, 11 corresponds to j3 = 2, 3, 4, the attitude subsystem shown in formula (9b) is considered as follows:

[0151]

[0152] The parameters used by the attitude subsystem and the position subsystem are distinguished by different subscripts. The controllers are different because the attitude subsystem shown in formula (39) is different from the position subsystem model shown in formula (23).

[0153] The state transformation is defined as shown in formula (40):

[0154]

[0155] in, is the new state quantity; Tracking error The error conversion, It is the virtual control input.

[0156] Select the Lyapunov function as follows:

[0157]

[0158] By using formula (15) and formula (40), we can get The derivative of :

[0159]

[0160] Design the virtual control input as:

[0161]

[0162] in, and It is a nonlinear expression term generated in the process of presetting performance method. is the design parameter. Substituting formula (43) into formula (42), we get:

[0163]

[0164] Select the Lyapunov function as follows:

[0165]

[0166] right The derivative is:

[0167]

[0168] Select the Lyapunov function as follows:

[0169]

[0170] right The derivative is:

[0171]

[0172] in, represents the design parameters, is an unknown positive constant and satisfies There is an unknown positive constant satisfy

[0173] Design the attitude subsystem controller And the fuzzy adaptive law is as follows:

[0174]

[0175] in, and is the design parameter, and Substituting formula (49) and formula (50) into formula (48), we obtain:

[0176]

[0177] According to Young's inequality, the following inequality holds:

[0178]

[0179] Substituting formula (52) and formula (53) into formula (51), we obtain:

[0180]

[0181] in,

[0182] According to the obtained coupling between the position virtual controller and the virtual input signal as shown in formula (33), the expression of the actual total lift u1 is as follows:

[0183]

[0184] The desired pitch angle and roll angle are obtained as follows:

[0185]

[0186] In the step 2, after the design of the preset performance controller of the quad-rotor UAV is completed, a stability analysis is performed on the UAV controlled by the preset performance controller of the UAV.

[0187] Specifically, after completing the design of the preset performance controller of the UAV, the stability analysis of the quadrotor UAV controlled by the controller is performed. For example, the quadrotor UAV system considering time-varying position constraints, model uncertainty, actuator failure and external wind disturbance as shown in formulas (9a) and (9b) is assumed to be valid. If the virtual control signals shown in formulas (27) and (43) and the actual controllers of the fuzzy adaptive law shown in formulas (34) and (50) are applied, the actual controllers are the controllers shown in formulas (33) and (49) respectively. For any initial condition, Tracking Error Ultimately, uniform bounds are guaranteed, where i4=1,2,3,4,5,6.

[0188] The following is a detailed description of the process of stability analysis of a quadrotor drone controlled by a drone preset performance controller:

[0189] a, b, c, d are real numbers. Let a∈R, b∈R, c>1 and d>1, and satisfy (c-1)(d-1)=1. For any k>0, the following inequality can be obtained:

[0190]

[0191] Select the overall Lyapunov function Where i3 = 1, 3, 5, 7, 9, 11; according to formula (38) and formula (54), the derivative of V is shown in formula (57):

[0192]

[0193] in,

[0194] By solving inequality (57), we obtain the inequality shown in formula (58):

[0195]

[0196] Where V(0) represents the value of V in the zero state, and p represents all The sum of ; this means that all closed-loop signals are bounded. In particular, according to formula (24), formula (40), formula (29) and formula (45), we get is bounded, and ε, is also bounded. Therefore, the tracking error z(t) is bounded and remains within the preset performance boundary, that is, the control system converges to meet the preset performance requirements. For the position subsystem, taking the x subsystem as an example, note that z1=Y1-x1, we have:

[0197]

[0198] It is expressed as:

[0199] Q1z1=Q2e1 (59)

[0200] Among them, when x∈D1, Q1 and Q2 are bounded functions, Q1=(F 11 +x d )(F 12 -x d )(F 11 +x)(F 12 -x)>0, Q2=F 11 F 12 +xx d >0; e1 is the actual tracking error, e1=x d -x. Therefore, formula (59) is rewritten as shown in formula (60):

[0201] e1=Qz1 (60)

[0202] in Where Q is a bounded function, because Q and z1 are bounded, so e1 is also bounded. Similar to the x subsystem, the actual tracking errors of other subsystems are also bounded.

[0203] Therefore, all closed-loop signals are bounded. Actual position state [x,y,z] T It can stably follow the given desired signal without violating the position state constraints.

[0204] Without loss of generality, the constraints in the present invention are asymmetric. When considering symmetric constraints, that is, As a known positive function, the designed controller can also ensure that the control requirements are met.

[0205] In summary, considering the faulty quadrotor drone model with time-varying position constraints, model uncertainty and external wind disturbance, the virtual control law shown in formula (27) and formula (43), the fuzzy adaptive law of formula (34) and formula (50) and the actual control signal shown in formula (33) and formula (49) are adopted. The tracking error will eventually converge to a small neighborhood of the origin. In addition, all signals in the closed-loop system are bounded.

[0206] Step 3. Use the preset UAV controller designed in step 2 to realize trajectory tracking control of the mapped unconstrained quadrotor UAV system, and use the mapping relationship to obtain the desired signal that the original constraint system can stably maintain stable following under the constraint conditions, that is, use the mapping relationship to obtain the desired signal that the original constraint system can stably maintain stable following under the constraint conditions.

[0207] In addition, in order to verify the effectiveness of the method proposed in the present invention, the following specific experiments are also given:

[0208] This embodiment will verify the effectiveness of the nonlinear mapping method for the original system shown in formula (1a) and formula (1b) through numerical simulation, as well as the effectiveness of the preset performance controller designed above for the state control of the drone system shown in formula (9a) and formula (9b) after preprocessing. In addition, in order to highlight the advantages of the proposed method, the adaptive inversion control ABC method of the original system is used to compare with the method proposed in the present invention.

[0209] The proposed UAV preset performance control method considering time-varying position constraints is simulated in a virtual environment to verify the feasibility of the control method proposed in the present invention:

[0210] The parameters of the drone are:

[0211]

[0212] The reference trajectory is chosen as:

[0213]

[0214] Give the initial value of X0:

[0215]

[0216] Among them, (x0, y0, z0) is the initial value of the position of the UAV in the inertial coordinate system, φ0, θ0, ψ0 represent the initial values ​​of the roll angle, pitch angle and yaw angle of the UAV respectively.

[0217] The parameters of the preset performance function are:

[0218]

[0219] Select the constraint boundary function as:

[0220]

[0221] The model uncertainties are:

[0222]

[0223] The external wind disturbances are:

[0224]

[0225] i4=1,2,3,4,5,6, the parameters of the drone preset performance controller are:

[0226]

[0227] In the simulation of the method of the present invention, the actuator fault parameters are as follows: Figure 4 As shown, the control input time response is Figure 5 The three-dimensional motion trajectory of the quadcopter is shown in Figure 6 shown.

[0228] Compared with the ABC method, Figure 7 The time response of the actual position trajectory (x, y, z) is shown, and some information can be obtained. Although both methods keep the position state within the constraint boundary, the control accuracy of the ABC method is obviously weaker than that of the method of the present invention. Figure 8 Middle x 、z y 、z z Respectively represent the tracking errors of the position subsystem x, y, and z. Figure 8 It can be seen that although the error of the method of the present invention is relatively obvious at the beginning of the simulation, it then quickly converges to a very small neighborhood of zero. Until the end of the simulation, the error of the ABC method is also relatively large, which shows that the error of the ABC method has a larger fluctuation than that of the method of the present invention as a whole.

[0229] The converted position tracking is as follows Fig. 9 As shown in Figure 1, it can be easily seen that the system state quickly and stably follows the reference signal. The time response of the position error after conversion is as follows: Fig.10 As shown, where z x1、z x3 、z x5 They represent the tracking errors of the conversion states of the position subsystem x, y, and z after mapping, and the errors eventually converge to zero.

[0230] The time response and error of posture tracking are respectively Fig.11 and Fig.12 As shown. Fig.11 It can be seen that the tracking performance of the method of the present invention is better than that of the ABC method. Fig.12 Middle φ 、z θ 、z ψ They represent the tracking errors of the roll angle, pitch angle and yaw angle of the attitude subsystem respectively. Compared with the ABC method, the method of the present invention has Fig.12 The error in the control method is smaller. Specifically, in the φ and θ subsystems, the error of the ABC method exceeds the preset boundary, while the method proposed in the present invention always remains within the preset boundary. In addition, similar to the position subsystem, until the end of the simulation, the error of the ABC method is greater than the error of the control method of the present invention. Through the above analysis, it is easy to obtain the information that the method of the present invention is more effective and faster than the ABC method in the attitude subsystem.

[0231] In order to further demonstrate the superiority of the method proposed in the present invention, the following performance indicators between the method proposed in the present invention and ABC are compared to further prove the effectiveness of the method proposed in the present invention.

[0232] Absolute Error (IAE) amplitude integral:

[0233]

[0234] Among them, i4=1,2,3,4,5,6.

[0235] x y z φ θ PM 0.1890 0.3599 0.0885 0.0015 0.0017 ABC 0.0959 0.3958 0.1055 0.0240 0.0213

[0236] Table 1

[0237] The IAE indicators of the PM and ABC methods of the present invention are shown in Table 1. Except for the x subsystem, the indicators of the method proposed in the present invention are smaller than those of the ABC method. d =0, we only need to study the indicators of x, y, z, φ, and θ.

[0238] Through the above analysis, it is concluded that under the conditions of actuator failure and external interference, the control method proposed in the present invention has good tracking performance, and the tracking error is kept within the preset performance range. The control method proposed in the present invention is significantly better than the tracking performance of the traditional ABC method.

[0239] Example 2

[0240] This embodiment 2 describes a quadrotor drone preset performance control system considering time-varying position constraints, which is based on the same inventive concept as the quadrotor drone preset performance control considering time-varying position constraints in embodiment 1.

[0241] The preset performance control system of the quadrotor drone considering time-varying position constraints includes the following modules:

[0242] The model building module is used to establish a composite model of the UAV that takes into account time-varying position constraints, model uncertainties, and external wind disturbances, and to map the composite model using a nonlinear mapping relationship to convert it into an unconstrained model; to limit the tracking error of the unconstrained system using a preset performance function, and to process the tracking error using an error transformation function, and to use it in the design of a subsequent preset performance controller for the UAV;

[0243] The controller design module is used to estimate model uncertainty based on an unconstrained system using an adaptive fuzzy system, estimate and compensate for external wind disturbances and actuator failures using a disturbance observer, and design a UAV preset performance controller using the preset performance method and backstepping method;

[0244] And an actuator fault control module is used to use the UAV preset controller to realize trajectory tracking control of the unconstrained system obtained after nonlinear mapping, and according to the nonlinear mapping relationship, the expected signal for the original constrained system to stably follow without violating the constraint conditions can be obtained.

[0245] It should be noted that in the system described in this embodiment, the implementation process of the functions and effects of each functional module is specifically described in the implementation process of the corresponding steps in the method in Example 1, and will not be repeated here.

[0246] Example 3

[0247] This embodiment 3 describes a computer device, which includes a memory and one or more processors. The memory stores executable code, and when the processor executes the executable code, it is used to implement the steps of the method for controlling the preset performance of a drone considering time-varying position constraints in the above embodiment 1.

[0248] In this embodiment, the computer device is any device or apparatus with data processing capability, which will not be described in detail here.

[0249] Example 4

[0250] This embodiment 4 describes a computer-readable storage medium, on which a program is stored, and when the program is executed by a processor, it is used to implement the steps of the method for controlling the preset performance of a drone considering time-varying position constraints in embodiment 1. The computer-readable storage medium can be an internal storage unit of any device or apparatus with data processing capabilities, such as a hard disk or memory, or an external storage device of any device with data processing capabilities, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), an SD card, a flash card (Flash Card), etc. equipped on the device.

[0251] Of course, the above description is only a preferred embodiment of the present invention, and the present invention is not limited to the above embodiments. It should be noted that all equivalent substitutions and obvious deformation forms made by any technician familiar with the field under the guidance of this specification fall within the essential scope of this specification and should be protected by the present invention.

Claims

1. A method for controlling the preset performance of a UAV considering time-varying position constraints, characterized in that: The steps include: Step 1. Establish a composite model of the UAV that takes into account time-varying position constraints, model uncertainty, and external wind disturbances, and use nonlinear mapping relationships to map the composite model and convert it into an unconstrained model; use a preset performance function to limit the tracking error of the unconstrained system, use an error transformation function to process the tracking error, and use it for the design of the subsequent preset performance controller of the UAV; Step 2. Based on the unconstrained system in step 1, the model uncertainty is estimated using an adaptive fuzzy system, and the external wind disturbance and actuator failure are estimated and compensated using a disturbance observer. The preset performance method and backstepping method are used to design the UAV preset performance controller. Step 3. Use the preset controller of the UAV designed in step 2 to realize trajectory tracking control of the unconstrained system obtained after nonlinear mapping, and according to the nonlinear mapping relationship, the expected signal of the original constrained system to stably follow without violating the constraints can be obtained.

2. The method for controlling the preset performance of a drone considering time-varying position constraints according to claim 1, characterized in that: In step 1, the process of establishing a composite model of the UAV considering time-varying position constraints, model uncertainty and external wind disturbances is as follows: Assume that the position vector P and the Euler angle vector θ are expressed as: P=[x,y,z] T ∈R 3 and Θ = [φ,θ,ψ] T ∈R 3 ; Where (x, y, z) is the position of the drone in the inertial coordinate system, φ, θ, ψ represent the roll angle, pitch angle and yaw angle of the drone respectively; let [P1(t), P2(t), P3(t)] T =[x,y,z] T ; Establish the UAV position subsystem model as shown in formula (1a) and the UAV attitude subsystem model as shown in formula (1b): Where A=[A1,A2,A3] T , A1=(cosφsinθcosψ+sinφsinψ) / m, A2=(cosφsinθsinψ-sinφcosψ) / m, A3=cosφsinθ / m, m represents the total mass of the drone; H a represents the unknown nonlinear composite term acting on the position subsystem, H a =[H1,H2,H3] T ∈R 3 , where H1, H2, and H3 represent the unknown nonlinear composite terms acting on the x-axis, y-axis, and z-axis directions of the position subsystem, respectively; d a (t) represents the external wind disturbance to the position subsystem, d a (t) = [d1, d2, d3] T ∈R 3 , where d1, d2, and d3 represent the external wind disturbances to the position subsystem in the x-axis, y-axis, and z-axis directions, respectively; J x , J y , J z They represent the UAV's moment of inertia about the x-axis, y-axis, and z-axis respectively, l is the distance from the UAV propeller to the center of mass; u Θ =[u o,2 ,u o,3 ,u o,4 ] T ∈R 3 ,u o,1 、u o,2 、u o,3 、u o,4 They represent the control inputs generated by the four motor rotors of the drone; H b represents the unknown nonlinear composite term acting on the attitude subsystem, H b =[H4,H5,H6] T ∈R 3 , where H4, H5, and H6 represent the unknown nonlinear composite terms acting on the roll angle, pitch angle, and yaw angle of the attitude subsystem, respectively; d b (t) represents the external wind disturbance to the attitude subsystem, d b (t) = [d4, d5, d6] T ∈R 3 , where d4, d5, and d6 represent the external wind disturbances on the roll angle, pitch angle, and yaw angle of the attitude subsystem, respectively; The UAV position subsystem and attitude subsystem models have the influence of asymmetric time-varying position constraints as shown in formula (2): Among them, i1=1,2,3, Represents a collection, is a known positive function; H (·) =f (·) +△f (·) ; Among them, H (·) represents the unknown nonlinear composite term acting on the position subsystem and attitude subsystem; f (·) represents the air resistance term, where f1, f2, f3, f4, f5, and f6 represent the air resistance in the x-axis, y-axis, z-axis, roll angle, pitch angle, and yaw angle directions, respectively; g is the acceleration due to gravity; G x , G y , G z , G φ , G θ , G ψ Respectively represent the air resistance coefficient of the drone in the x-axis, y-axis, z-axis, roll angle, pitch angle and yaw angle directions; △f () represents model uncertainty, [△f1,△f2,△f3] T ∈R 3 and [△f4,△f5,△f6] T ∈R 3 Respectively represent the model uncertainties of the position subsystem and attitude subsystem; u o,1 、u o,2 、u o,3 、u o,4 The relationship is: Among them, l1 is the lift coefficient, l2 is the rotor drag coefficient, w1, w2, w3, w4 are the motor rotor speeds; To avoid singularity problems, the roll angle and pitch angle should satisfy and The external wind disturbance is unknown and bounded; the time-varying position constraint boundaries do not cross, that is, the upper constraint bound is always greater than the lower constraint bound; Let the expected trajectory [Υ1,γ2,Υ3] T =[x d ,y d ,z d ] T ; where x d ,y d 、z d Represents the expected position trajectory of the drone in the inertial coordinate system; there is a positive constant and Make the expected trajectory and its derivative satisfies and And for the desired yaw angle ψ d Just meet Where j1=1,2; In addition, the system states x, y, z, θ, ψ, and φ can be used to design the controller for the preset performance of the UAV; When the four actuators of the drone are affected by a fault during operation, the actual input With expected input There are differences, considering the relationship shown in formula (4): Among them, i2=1,2,3,4, and is a continuous time-varying function, and represent the partial fault factor and bias fault factor respectively, and and are all unknown and bounded continuous functions.

3. The method for controlling the preset performance of a drone considering time-varying position constraints according to claim 2, characterized in that: In step 1, the composite model is mapped using a nonlinear mapping relationship to convert it into an unconstrained model as follows: In a nonlinear system, assume that the constraint state is s, the mapped new state is set to ζ, and the constraint range is -k c1 <s<k c2 , where k c1 , k c2 are all positive numbers; consider the smooth nonlinear mapping expression N(k c1 ,k c2 ,s):s→ζ, where N(·) is a nonlinear mapping function and the expression of ζ is shown in formula (5): When s approaches the constraint boundary, ζ→∞; the value range of the mapped new state ζ is unconstrained, that is, ζ∈R; Defining system status And suppose [x1,x3,x5] T The actual position state [x,y,z] T The conversion state obtained by the nonlinear mapping function N(·), that is, x1=N(F 11 ,F 12 ,x),x3=N(F 21 ,F 22 ,y),x5=N(F 31 ,F 32 ,z); through the transformation of nonlinear mapping function, we get [x1,x3,x5] T ∈R 3 ; In order to obtain the transformed unconstrained model, Taking the derivative we get: in and The expression of is shown in formula (7), The expression formula (8) is shown as follows: In the previously defined collection middle, is a positive number and can be calculated; combined with the UAV dynamics model shown in formula (1a) and formula (1b), the time-varying position constraint shown in formula (2), the actuator fault shown in formula (4) and the nonlinear mapping function N(·), the UAV position subsystem model S1 of the unconstrained system shown in formula (9a) and the UAV attitude subsystem model S2 of the unconstrained system shown in formula (9b) are obtained respectively: Among them, [x1,x3,x5] T is the unconstrained state after the transformation; u x 、u y and u z Represent the virtual control inputs of the x-subsystem, y-subsystem, and z-subsystem, respectively, and u x =u1(cosφsinθcosψ+sinφsinψ),u y =u1(cosφsinθsinψ-sinφcosψ),u z =u1cosφcosθ; D (·) represents the nonlinear term containing disturbance, where D1, D2, D3, D4, D5, and D6 represent the nonlinear terms containing disturbance in the x-axis, y-axis, z-axis, roll angle, pitch angle, and yaw angle directions, respectively. D1=d1+A1o1-A1Γ1u1, D2=d2+A2o1-A2Γ1u1, D3=d3+A3o1-A3Γ1u1, D4=d4+K2o2-K2Γ2u2, D5=d5+K3o3-K3Γ3u3, D6=d6+K4o4-K4Γ4u4; D (·) depends on external wind disturbance, system state, fault parameters and expected input; the system state depends on the control input, and the control input that drives the drone depends on the speed of the four rotors of the drone. The speed derivatives of the four rotors are bounded and there is a positive constant Make The position subsystem shown in formula (9a) is a nonlinear mapping subsystem, and the position controller is designed based on the unconstrained state; therefore, the desired position signal also needs to be mapped as shown in formula (10): [Y1,Y2,Y3] T represents the expected position signal after mapping processing; Assume that the expected gesture signal is [Y4, Y5, Y6] T =[φ d ,θ d ,ψ d ] T And used in the design of subsequent UAV preset performance controller, where φ d ,θ d , d Respectively represent the desired roll angle, expected pitch angle and expected yaw angle of the drone; at the same time, define the position and rotation tracking error for: Among them, i3=1,3,5,7,9,11; When i1=1,2,3, for any If for any t, are all bounded, then it can be guaranteed that for any t, Where t represents time, represents the initial position of x, y, z; the position constraint problem is then reduced to ensuring The boundedness problem; In step 1, the tracking error of the unconstrained system is limited by using a preset performance function, and the tracking error is processed by using an error transformation function as follows: The preset performance method is shown in formula (12): -ρ(t) <z(t)<ρ(t) (12) Wherein, z(t) represents the tracking error signal of the unconstrained system shown in formula (9a) and formula (9b), ρ(t) is the preset performance function, and the preset performance function is selected as shown in formula (13): ρ(t)=(ρ0-ρ ∞ )e -ht +r ∞ (13) Among them, ρ0, ρ ∞ , h is a positive constant, and satisfies ρ0>ρ ∞ ; In order to achieve the preset performance control of tracking error, the error transformation function G(ε) is introduced: z(t)=ρ(t)G(ε) (14) Among them, G(ε) is a differentiable function with positive derivative, ε is the new conversion error, and the expression of ε is: Taking the derivative of the new conversion error ε, we get: in and Through conversion error processing, the tracking error z(t) is ensured to be within the preset performance requirement range, proving that the conversion error ε is bounded.

4. The method for controlling the preset performance of a drone considering time-varying position constraints according to claim 3, characterized in that: The step 2 is specifically as follows: Step 2.

1. Design a fuzzy observer consisting of an adaptive fuzzy system and a disturbance observer; For composite items Perform fuzzy approximation and get: in, represents an adaptive fuzzy system, represents the weight, represents the Gaussian basis function, represents the approximation error; Design a fuzzy observer to approximate the disturbance compound term and define Let be a bounded composite term containing perturbations. Definition o =x,y,z,2,3,4 and K o =[K1,K1,K1,K2,K3,K4] T Corresponding to i3=1,3,5,7,9,11 respectively, the fuzzy observer is designed as shown in formula (18): in, yes The estimate, and are the design parameters and observer states, respectively, and yes estimates; right The derivative is: in, express The estimation error of set up for The estimation error of is obtained according to formula (19): Applying Young's inequality to the last term in equation (20) yields: in, is an unknown positive constant and satisfies There is an unknown positive constant satisfy Substituting formula (21) into formula (20), we obtain: in, Step 2.

2. Design the UAV preset performance controller using the preset performance method and backstepping method; When i5 = 1, 3, 5 correspond to j2 = x, y, z respectively, consider the position subsystem shown in formula (9a), as shown in formula (23): Define the state transformation as shown in formula (24): in, is the new state quantity; Tracking error The error conversion, is the virtual control input; Select the Lyapunov function shown in formula (25) By using formula (15) and formula (24), we can get The derivative of is: in, and It is a nonlinear expression term generated in the process of presetting performance method; Design the virtual control input as: in is the design parameter. Substituting formula (27) into formula (26), we get: Choose the Lyapunov function as shown in formula (29) right The derivative is: in, Choose the Lyapunov function as shown in formula (31) right The derivative is: in, represents the design parameters, is an unknown positive constant and satisfies There is an unknown positive constant satisfy Design the position subsystem controller as shown in formula (33) Design the fuzzy adaptive law as shown in formula (34): in, and is the design parameter, and Substituting formula (33) and formula (34) into formula (32), we obtain: According to Young's inequality, the inequalities of formula (36) and formula (37) hold: Substituting formula (36) and formula (37) into formula (35), we obtain: in, When i6=7,9,11 corresponds to j3=2,3,4, the attitude subsystem shown in formula (9b) is considered as shown in formula (39): The state transformation is defined as shown in formula (40): in, is the new state quantity; Tracking error The error conversion, is the virtual control input; Select the Lyapunov function shown in formula (41) By using formula (15) and formula (40), we can get The derivative of is: Design the virtual control input as: in, and It is a nonlinear expression term generated in the process of presetting performance method. is the design parameter. Substituting formula (43) into formula (42), we get: Choose the Lyapunov function as shown in formula (45) right Taking the derivative we get: Choose the Lyapunov function as shown in formula (47) right Taking the derivative we get: in, represents the design parameters, is an unknown positive constant and satisfies There is an unknown positive constant satisfy Design the attitude subsystem controller as shown in formula (49): And the fuzzy adaptive law shown in formula (50): in, and is the design parameter, and Substituting formula (49) and formula (50) into formula (48), we obtain: According to Young's inequality, the inequalities of formula (52) and formula (53) hold: Substituting formula (52) and formula (53) into formula (51), we obtain: in, According to the obtained coupling between the position virtual controller and the virtual input signal as shown in formula (33), the expression of the actual total lift u1 is shown in formula (55): Get the desired pitch angle θ d and the desired roll angle φ d As shown in formula (56):

5. The method for controlling the preset performance of a drone considering time-varying position constraints according to claim 1, characterized in that: In the step 2, after the design of the drone preset performance controller is completed, a stability analysis is performed on the drone controlled by the drone preset performance controller.

6. The method for controlling the preset performance of a drone considering time-varying position constraints according to claim 4, characterized in that: For the new unconstrained system after mapping, its dynamic model is shown in formula (9a) and formula (9b), considering the quadratically differentiable expected tracking signal Y i If the virtual control signals shown in formula (27) and formula (43) and the fuzzy adaptive laws shown in formula (34) and formula (50) are used, and the actual control signals shown in formula (33) and formula (49) and the fuzzy observer shown in formula (18) are used, the trajectory tracking error of the UAV system after mapping processing will be guaranteed to be uniformly bounded, and the expected signal of the original constraint system that can stably follow without violating the constraint conditions can be obtained according to the nonlinear mapping relationship.

7. The method for controlling the preset performance of a drone considering time-varying position constraints according to claim 2, characterized in that: In step 1, when o i2 = 0, the actuator only has a bias fault, and when o i2 When ≠0, the actuator has a bias and gain fault.

8. A UAV preset performance control system considering time-varying position constraints, characterized in that: Includes the following modules: The model building module is used to establish a composite model of the UAV that takes into account time-varying position constraints, model uncertainties, and external wind disturbances, and to map the composite model using a nonlinear mapping relationship to convert it into an unconstrained model; to limit the tracking error of the unconstrained system using a preset performance function, and to process the tracking error using an error transformation function, and to use it in the design of a subsequent preset performance controller for the UAV; The controller design module is used to estimate model uncertainty based on an unconstrained system using an adaptive fuzzy system, estimate and compensate for external wind disturbances and actuator failures using a disturbance observer, and design a UAV preset performance controller using the preset performance method and backstepping method; And an actuator fault control module is used to use the UAV preset controller to realize trajectory tracking control of the unconstrained system obtained after nonlinear mapping, and according to the nonlinear mapping relationship, the expected signal for the original constrained system to stably follow without violating the constraint conditions can be obtained.

9. A computer device comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that: When the processor executes the executable code, the steps of the method for controlling the preset performance of a drone considering time-varying position constraints as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by the processor, the steps of the method for controlling the preset performance of a drone taking into account time-varying position constraints as described in any one of claims 1 to 7 are implemented.

Images