An unmanned aerial vehicle active fault-tolerant control method with jitter suppression performance

Abstract

The application discloses a kind of unmanned aerial vehicle active fault-tolerant control methods with jitter suppression performance, it is related to unmanned aerial vehicle control technical field.The technical points of the present application include: establishing linear discrete-time state space expression form unmanned aerial vehicle model;Using hidden half Markov switching system theory to model inaccurate FDI process of observation information, build the anti-shake controller dependent on observation mode;Unmanned aerial vehicle system state anti-shake constraint condition is built;Propose the sufficient condition for guaranteeing the mean square stability of unmanned aerial vehicle system under inaccurate FDI information condition;Propose the sufficient condition for the existence of fault-tolerant controller of discrete-time unmanned aerial vehicle system with anti-shake performance;Matrix inequality is solved, and unmanned aerial vehicle system fault-tolerant controller gain is obtained;Unmanned aerial vehicle system fault-tolerant controller gain is used to realize active fault-tolerant control to quad-rotor unmanned aerial vehicle.The controller obtained by the application can simultaneously consider the state jitter suppression performance of system and system stability, improve the flight stability and safety of unmanned aerial vehicle.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control technology, and specifically to an active fault-tolerant control method for UAVs with jitter suppression capabilities. Background Technology

[0002] In UAV control systems, external disturbances, high-maneuverability mission requirements, and system switching behaviors can all induce frequent and severe jitter in the UAV system's state. If no measures are taken to limit and constrain this jitter, it will accumulate, eventually causing damage to the controlled object or actuators, severely impacting the safety and reliability of the UAV system. Therefore, suppressing the state jitter behavior of UAV control systems, especially its dynamic characteristics, is crucial. It is worth noting that existing anti-jitter control strategies primarily target the jitter of control input signals. However, for state jitter, which also affects the transient performance of the UAV system, designing corresponding anti-jitter control strategies is more challenging and has received little research.

[0003] On the other hand, highly integrated UAVs may experience random motor failures due to factors such as motor overload, stall, and saturation during high-maneuverability flight missions. Existing fault-tolerant control methods often use stochastic processes such as Markov chains and Poisson distributions to describe the occurrence of such random failures, assuming that the fault detection and identification process—FDI information—is completely and accurately known. However, in engineering practice, considering factors such as sensor measurement noise, the obtained FDI information is not actually accurate, and sometimes it is even unavailable. Therefore, designing an effective modeling and fault-tolerant control method to eliminate the impact of inaccurate fault information on system stability has significant theoretical and engineering implications.

[0004] It is worth noting that while most existing research on fault-tolerant control uses stochastic processes such as Markov chains to model randomly occurring actuator failures, the resulting state jitter in the control system has received little attention. In reality, while the introduction of stochastic processes into the system accurately models random failures, their switching behavior can also induce significant changes in the system state. Although existing fault-tolerant control methods can mitigate or avoid the negative impacts of actuator failures on UAV control systems to some extent, the resulting state jitter still affects the reliability of the control system and the safety of the UAV system. This frequent and severe jitter also seriously affects the performance of the UAV control system, highlighting the shortcomings of existing fault-tolerant control and state jitter prevention control theories and methods. Summary of the Invention

[0005] In view of the above problems, this invention proposes an active fault-tolerant control method for unmanned aerial vehicles (UAVs) with jitter suppression performance, which includes the following steps:

[0006] Step 1: Establish a UAV model in the form of a linearized discrete-time state-space expression;

[0007] Step 2: Model the FDI process due to inaccurate observation information using the theory of hidden semi-Markov switching systems, and construct a jitter stabilization controller that depends on the observation mode; including:

[0008] A semi-Markov random process is used to model the observed FDI information, and a piecewise homogeneous time-varying emission probability is used to represent the probabilistic relationship between the observed FDI information and the true FDI information, thus establishing a piecewise homogeneous emission probability dominated by a Markov chain.

[0009] By combining the hidden semi-Markov model and the piecewise homogeneous emission probability, a state-based anti-shake controller that depends on the observation mode is designed.

[0010] Step 3: Construct the state stabilization constraints for the UAV system;

[0011] Step 4: Propose sufficient conditions to ensure the mean-square stability of the UAV system under conditions of inaccurate FDI information; including:

[0012] A lemma for determining the mean-square stability of a general switching system is given.

[0013] Combining the FDI process modeling method with inaccurate observation information in step two, we establish sufficient conditions to ensure the mean square stability of the UAV system when fault information is inaccurate or unavailable.

[0014] Step 5: Propose sufficient conditions for the existence of a fault-tolerant controller with anti-shake performance in a discrete-time UAV system; including:

[0015] The state anti-shake constraint condition of the UAV system in step 3 and the sufficient condition for the mean square stability of the UAV system in step 4 are converted into matrix inequalities. The matrix inequalities are combined to obtain the sufficient condition for the existence of a fault-tolerant controller with anti-shake performance.

[0016] Step 6: Solve the matrix inequalities obtained in Step 5 to obtain the gain of the fault-tolerant controller of the UAV system;

[0017] Step 7: Utilize the gain of the UAV system fault-tolerant controller to achieve active fault-tolerant control of the quadcopter UAV.

[0018] Furthermore, the drone model described in step one is represented as follows:

[0019] x(k+1)=A(η k )x(k)+B(η k )u(k)

[0020] Where k represents time; x(k) represents the UAV system state matrix; and u(k) represents the control input matrix. The positive constant s represents the sampling period of the system, and I represents the identity matrix; and Represents the system matrix. The uncertainty coefficient represents the model of the unmanned aerial vehicle. This represents the efficiency loss caused by fluctuations in the motor's output torque, and {η k Let} denote a stochastic process that follows a semi-Markov property, and this stochastic process lies in a finite set. Take the value from; A eq Representing the system matrix The matrix variable in B eq Representing the system matrix Matrix variables in; ΔA eq (η k ) represents an uncertain parameter with multimodal characteristics.

[0021] Furthermore, the expression for the piecewise homogeneous emission probability dominated by the Markov chain in step two is:

[0022]

[0023] in, Indicates the piecewise homogeneous time-varying emission probability; The observation mode representing FDI information; Represents the true mode of FDI information; σ k Let τ denote a high-dimensional Markov random chain; τ denote the dwell time; o(τ) denote the true mode of FDI information. any value in the FDI information, where 'a' represents the observation mode of the FDI information. An arbitrary value in σ, where α represents the high-dimensional Markov random chain σ. k The piecewise homogeneous time-varying emission probability is fixed within a time interval, and the switching between different time intervals is determined by a high-dimensional Markov chain {σ}. k}Dominate.

[0024] Furthermore, the structure of the state stabilization controller described in step two is as follows:

[0025]

[0026] in, It is the controller gain to be solved.

[0027] Furthermore, the state stabilization constraints of the UAV system described in step three are constructed as follows:

[0028]

[0029] Where ε is a given constant matrix and γ is a positive constant;

[0030] Furthermore, the lemma for determining the mean square stability of the general switching system in step four is as follows:

[0031] Consider discrete-time general switching systems with finite dwell times. The switching times are k0, k1, ..., kn, and k0 = 0; Represents a class of continuous-time functions;

[0032] If for and Where N is the set of integers. For the set of integers from kn to kn+1, there exists a series of Lyapunov functions. and three function: For any initial conditions and given finite constants Where N1 represents the set of integers to which the random process of fault occurrence belongs, and N2 represents the set of integers to which the random process of FDI belongs; we can obtain:

[0033]

[0034]

[0035]

[0036] Where E represents the expectation, then the switching system is mean-square stable.

[0037] Furthermore, based on the lemma for the mean square stability of general switching systems in step four, the sufficient condition for the mean square stability of the UAV dynamic system in the absence of accurate FDI information is as follows:

[0038] Considering the unmanned aerial vehicle control system and finite constants If for There exists a matrix set Q ia (τ,v)>0,R ia (h,m)>0, and for The following inequalities must be satisfied:

[0039]

[0040]

[0041]

[0042]

[0043] The unmanned aerial vehicle control system It is mean-square stable;

[0044] in, This represents the maximum residence time of the system when the system mode is i, which is within the set of positive integers N. + The value is taken from τ; h represents any value in the set of integers N; A(τ-v-1) is an intermediate process variable;

[0045] Furthermore, the sufficient condition for the existence of a fault-tolerant controller with anti-shake performance in step five is:

[0046] Considering the unmanned aerial vehicle control system and finite constant ε ia >0, γ>0, a∈N2, finite matrix ε;

[0047] If for Existing matrix set And for The following inequalities must be satisfied:

[0048]

[0049]

[0050]

[0051]

[0052]

[0053] in,

[0054]

[0055]

[0056]

[0057] At this time, the drone control system It is mean-square stable and has a given anti-shake performance; wherein, the allowable controller gain is Let be the set positive definite matrix.

[0058] The beneficial technical effects of this invention are:

[0059] To mitigate the state jitter caused by frequent and significant changes in system state during UAV maneuvers, this invention proposes a stabilization controller design method with guaranteed anti-jitter performance. This method simultaneously ensures dynamic system stability and anti-jitter performance, effectively preventing phenomena such as airframe jitter, UAV system instability, and even damage to actuators caused by drastic changes in system state and actuator failures. Furthermore, addressing the problem of inaccurate observation of random fault information in UAV actuators, this invention proposes a UAV fault detection and identification process modeling method based on Hidden Semi-Markov Switched System Theory. This method describes the relationship between observed fault information and actual fault information using a piecewise homogeneous launch probability. By designing an observation mode-dependent controller, the stability of the UAV control system is guaranteed even when actuator fault information cannot be detected or is detected inaccurately. This method has significant theoretical and engineering practical value.

[0060] This invention proposes state stabilization constraints for UAVs, enabling the obtained controller to simultaneously consider the system's state jitter suppression performance and system stability, thereby improving the flight stability and safety of UAVs.

[0061] This invention overcomes the ideal assumption in the prior art that the fault detection and identification process information is precisely known, and proposes a fault-tolerant controller design method based on the hidden semi-Markov model that depends on the observation mode, so that the obtained results have lower conservatism. Attached Figure Description

[0062] The present invention can be better understood by referring to the description given below in conjunction with the accompanying drawings, which together with the following detailed description are included in and form part of this specification, and are used to further illustrate preferred embodiments of the invention and explain the principles and advantages of the invention.

[0063] Figure 1 This is a schematic diagram of a system structure with actuator failure and anti-jitter constraints.

[0064] Figure 2 The system state response diagrams are shown under different image stabilization performances, including image stabilization performance parameters γ = 5, 10, 20; the horizontal axis in the diagram represents the sampling time, and the vertical axis represents the system state [θp].

[0065] Figure 3 The diagram shows the open-loop and closed-loop state response of the UAV system. In the three-dimensional diagram above, the solid and dashed lines represent the closed-loop and open-loop systems, respectively, and the black and dark gray lines represent the states [φψq] and [pθr] (Euler angle vectors and body angular velocity vectors of the three coordinate axes of the UAV), respectively. In the two-dimensional diagram below, the horizontal axis represents the sampling time k, and the vertical axis represents the system modes.

[0066] Figure 4 The diagram shows the sequence of observation modes of the UAV under different launch probabilities. The horizontal axis represents the sampling time k, and the vertical axis represents the FDI observation mode. The first diagram corresponds to the first launch probability, the second diagram corresponds to the second launch probability, and the third diagram corresponds to the third launch probability. When the system's observation mode and the actual mode are exactly the same, they are marked with a light color during that sampling time. Detailed Implementation

[0067] To enable those skilled in the art to better understand the present invention, exemplary embodiments or examples of the present invention will be described below in conjunction with the accompanying drawings. Obviously, the described embodiments or examples are merely some, not all, of the embodiments or examples of the present invention. All other embodiments or examples obtained by those skilled in the art based on the embodiments or examples of the present invention without inventive effort should fall within the scope of protection of the present invention.

[0068] This invention addresses the state stabilization and fault-tolerant control problems of unmanned aerial vehicles (UAVs). It establishes a mean square stability criterion for hidden semi-Markov switching systems based on linear matrix inequalities and proposes a design method for an active fault-tolerant controller for UAVs with jitter suppression performance. This ensures the mean square stability of the UAV system while also providing good state stabilization performance.

[0069] An active fault-tolerant control method for UAVs with jitter suppression performance is implemented according to the following steps:

[0070] Step 1: Establish a UAV model in the form of a linearized discrete-time state-space expression;

[0071] Step 2: Model the inaccurate FDI process using Hidden Semi-Markov Switching System Theory and construct a state-dependent anti-jitter controller. The specific steps are as follows: First, use a semi-Markov stochastic process to model the observed FDI information, and represent the probabilistic relationship between the observed FDI information and the true FDI information using piecewise homogeneous time-varying emission probabilities. Second, establish a piecewise homogeneous emission probability governed by a Markov chain. Finally, combine the Hidden Semi-Markov model and the piecewise homogeneous emission probabilities to design a state-dependent anti-jitter controller.

[0072] Step 3: Construct state stabilization constraints for the UAV system;

[0073] Step 4: Propose sufficient conditions to ensure the mean square stability of the UAV system under inaccurate FDI information. The specific steps are as follows: First, provide a lemma for determining the mean square stability of a general switching system; second, combining the FDI process modeling method with inaccurate observation information from Step 2, establish sufficient conditions to ensure the mean square stability of the UAV system under inaccurate or unavailable fault information.

[0074] Step 5: Propose sufficient conditions for the existence of a fault-tolerant controller with anti-shake performance for a discrete-time UAV system. The specific steps are as follows: First, convert the anti-shake constraint condition of the UAV system state in Step 3 and the sufficient condition for the UAV system's mean-square stability in Step 4 into matrix inequalities using inequality scaling and other methods; second, simultaneously solve the above matrix inequalities to obtain sufficient conditions for the existence of a fault-tolerant controller with anti-shake performance.

[0075] Step 6: Solve the matrix inequalities obtained in Step 5 to obtain the gain of the fault-tolerant controller of the UAV system;

[0076] Step 7: Utilize the gain of the UAV system fault-tolerant controller to achieve active fault-tolerant control of the quadcopter UAV. Specific Implementation Example 1:

[0078] Step 1: Establish a discrete-time quadcopter UAV model:

[0079] x(k+1)=A(η k )x(k)+B(η k u(k) (1)

[0080] in, The positive constant s is the system's sampling period. and The matrix represents the system status and control input of the unmanned aerial vehicle (UAV). and It is a system matrix. The uncertainty coefficient represents the model of the unmanned aerial vehicle. This represents the efficiency loss caused by fluctuations in the motor's output torque, and {η k Let} denote a stochastic process that follows a semi-Markov property, and this stochastic process lies in a finite set. Take the value from the middle.

[0081] Step 2: Model the FDI process with inaccurate observation information using the theory of hidden semi-Markov switching systems, and construct a jitter stabilization controller structure that depends on the observation mode;

[0082] The FDI process with inaccurate observation information is modeled using the theory of hidden semi-Markov switching systems. The observation mode representing FDI information, The true mode of FDI information is represented, and the probabilistic relationship between the two is expressed by the emission probability. Unlike the time-invariant emission probability in existing theories, the emission probability of this invention is time-varying and has piecewise homogeneous properties:

[0083]

[0084] Among them, the probability of emission It is segmented and homogeneous, and the dwell time is... It remains constant within a time interval, and the switching between different time intervals is caused by a high-dimensional Markov chain {σ}. k Domination, for Its transition probability is defined as

[0085]

[0086] and

[0087] The controller structure constructed in this invention is as follows:

[0088]

[0089] in This is the controller gain to be solved. Combining equations (1) and (4), the UAV control system can be further described as for...

[0090]

[0091] in

[0092] Step 3: Construct the state stabilization constraints for the UAV control system:

[0093] To quantify the anti-shake performance of a drone system, this invention proposes the following system state anti-shake constraints:

[0094]

[0095] Where ε is a given constant matrix and γ is a positive constant.

[0096] Step 4: Propose sufficient conditions to ensure the mean-square stability of the UAV system under the condition of inaccurate FDI information:

[0097] First, construct a Lyapunov function of the following form:

[0098]

[0099] Next, referring to Lemma 1: Consider a discrete-time general switching system with finite dwell time. The switching times are k0, k1, ..., kn, ... and k0 = 0. If for and There exists a series of Lyapunov functions and three function For any initial conditions and given finite constants We can obtain:

[0100]

[0101]

[0102]

[0103] Therefore, the switching system is mean-square stable. Based on this, the sufficient condition for the mean-square stability of the UAV dynamic system in the absence of accurate FDI information can be obtained as follows:

[0104] Theorem 1: Consider the unmanned aerial vehicle control system (5) and the finite constant ε ia >0, If i∈N1, a∈N2, if for There exists a matrix set Q ia (τ,v)>0,R ia (h,m)>0, and for The following inequalities must be satisfied:

[0105]

[0106]

[0107]

[0108]

[0109] in: Recorded as The unmanned aerial vehicle system (5) is mean square stable.

[0110] Proof: Based on the constructed Lyapunov function The formula can be obtained

[0111]

[0112] in for It can be obtained

[0113]

[0114] Furthermore, for From formula (11), we can deduce

[0115]

[0116] Combining formulas (12) and (17), the following inequality can be derived:

[0117]

[0118] It can be seen that a positive definite matrix is ​​defined. and Formula (16) can then be derived from the above equation. When the system mode is changed by η k =i, to η k =j, At that time, it can be deduced

[0119]

[0120] From formula (19), we can deduce

[0121]

[0122] in, and By definition Formula (20) can be obtained from the following inequality.

[0123]

[0124]

[0125] From inequality (13), we know that for The following inequalities hold

[0126]

[0127] From formulas (18), (21)-(23), we can obtain Theorem 1, and the proof is complete.

[0128] Step 5: Establish a fault-tolerant controller with anti-shake performance for the UAV control system. (Conditions for its existence are listed below.)

[0129] Theorem 2: Consider the unmanned aerial vehicle system (5) and the finite constant ε ia >0, γ>0, i∈N1, a∈N2, finite matrix ε. If for Existing matrix set And for The following inequalities must be satisfied:

[0130]

[0131]

[0132]

[0133]

[0134]

[0135] in

[0136]

[0137]

[0138]

[0139]

[0140] At this point, the unmanned aerial vehicle system (5) is mean-square stable and has a given anti-shake performance. The allowable controller gain is...

[0141] Proof: By definition and Able to launch use replace At this point, formulas (11), (12), and (14) can be written as follows:

[0142]

[0143]

[0144]

[0145] From formula (13), we can obtain

[0146]

[0147] By Schur's complement lemma, the above equation is equivalent to

[0148]

[0149] in Considering matrix H ia (τ,v), L ia (τ,v) and Since they are all positive definite matrices, it can be deduced that...

[0150]

[0151]

[0152]

[0153] Furthermore, it can be deduced that...

[0154]

[0155]

[0156]

[0157] By definition and Inequality can be obtained and X a Make a contractual transformation, where X a It is a positive definite matrix. Further scaling using inequalities yields the following matrix inequalities.

[0158]

[0159]

[0160]

[0161]

[0162] Formula (6) can be derived

[0163]

[0164] Further, we can obtain

[0165]

[0166] Theorem 2 can be obtained from formulas (40), (41)-(43), and (45), thus completing the proof.

[0167] Solving the matrix inequality of Theorem 2 yields the gain of the fault-tolerant controller for the UAV system that meets the conditions, thereby realizing an active fault-tolerant control method for UAVs with jitter suppression performance.

[0168] Specific Implementation Example 2: This example illustrates a numerical simulation of an active fault-tolerant control method for unmanned aerial vehicles (UAVs) with jitter suppression capabilities. The simulation example and algorithm parameters are as follows: Assume the system has three switching modes, and parameters ζ1 = 1, ζ2 = 1.1, ζ3 = 0.9. The initial value of the system state is x(0) = [1 1 1 -5-5 -5] T .for set up

[0169]

[0170]

[0171]

[0172]

[0173] and Transition probability of FDI process Given

[0174] Table 1 shows the emission probability and transition probability matrices under different conditions.

[0175]

[0176] The dwell time probability density function for a fault is:

[0177]

[0178] The probability density function of the residence time in the FDI process is: And ρ ii (τ) = 0 (i = 1, 2, 3).

[0179] Substitute the above data into Theorem 2 obtained in step 5 to solve for the active fault-tolerant controller of the UAV with jitter suppression performance, and the simulation ends.

[0180] The simulation results of specific embodiment two are as follows: Figure 2 , Figure 3 , Figure 4 As shown. Under the condition of the transition probability matrix in condition 1, 50 state responses are generated, with the relevant parameters set to γ ​​= 5, 10, 20 and E = I. Subsequently, the average value of the 50 state responses is calculated, and the maximum and minimum values ​​of each response are recorded.

[0181] Figure 2 The average value and range of the Euler angle φ and Euler angular velocity p are described. It can be observed that the reachable boundary of the trajectory with parameter γ = 5 is significantly smaller, and the anti-shake performance is better than other cases. Specifically, the smaller the value of parameter γ, the smaller the state response range, and the better the anti-shake performance. The anti-shake controller designed according to the theory of this invention can effectively keep the UAV system stable even when the state changes drastically, demonstrating the effectiveness of the proposed anti-shake control method.

[0182] Figure 3 The state responses of the UAV open-loop and closed-loop systems under condition 1 are presented, where the anti-shake performance indicators are γ = 20 and E = I. Solid and dashed lines represent the closed-loop and open-loop systems, respectively, and black and dark gray lines represent states [φψq] and [pθr], respectively. Clearly, the state response of the closed-loop system eventually converges to 0, verifying the effectiveness of the fault-tolerant controller designed in this invention.

[0183] Figure 4 The observed mode sequences of the system are presented under three different launch probabilities. When the observed mode and the actual mode of the system are exactly the same, they are marked with a light color. It can be seen that the anti-shake control method proposed in this invention can effectively stabilize the UAV control system when the observation accuracy is 53%, 46%, and 32%, respectively.

[0184] Although the invention has been described with respect to a limited number of embodiments, those skilled in the art will understand from the foregoing description that other embodiments are conceivable within the scope of the invention described herein. The disclosure of the invention is illustrative and not restrictive, and the scope of the invention is defined by the appended claims.

Claims

1. An active fault-tolerant control method for unmanned aerial vehicles (UAVs) with jitter suppression performance, characterized in that, Includes the following steps: Step 1: Establish a UAV model in the form of a linearized discrete-time state-space expression; Step 2: Model the FDI process where observation information is inaccurate using the theory of hidden semi-Markov switching systems, and construct a jitter stabilization controller that depends on the observation mode; including: A semi-Markov random process is used to model the observed FDI information, and a piecewise homogeneous time-varying emission probability is used to represent the probabilistic relationship between the observed FDI information and the true FDI information, thus establishing a piecewise homogeneous emission probability dominated by a Markov chain. By combining the hidden semi-Markov model and the piecewise homogeneous emission probability, a state-based anti-shake controller that depends on the observation mode is designed. Step 3: Construct the following state stabilization constraints for the UAV system: in, Given a constant matrix, γ is a positive constant; k represents time. Represents the state matrix of the unmanned aerial vehicle (UAV) system; , The positive constant s represents the sampling period of the system. Represents the identity matrix; and Represents the system matrix. This represents the uncertainty coefficient of the drone model. This represents the efficiency loss caused by fluctuations in the motor's output torque, and ; Let represent a stochastic process that follows a semi-Markov property, and this stochastic process is within a finite set. Take the value from; Representing the system matrix Matrix variables in Representing the system matrix Matrix variables in; Represents an uncertain parameter exhibiting multimodal characteristics; This represents the controller gain to be solved. The observation mode representing FDI information; Represents the true modality of FDI information; Represents a high-dimensional Markov random chain; Step 4: Propose sufficient conditions to ensure the mean-square stability of the UAV system under conditions of inaccurate FDI information; including: A lemma for determining the mean-square stability of a general switching system is given. Combining the FDI process modeling method with inaccurate observation information in step two, we establish sufficient conditions to ensure the mean square stability of the UAV system when fault information is inaccurate or unavailable. Step 5: Propose sufficient conditions for the existence of a fault-tolerant controller with anti-shake performance in a discrete-time UAV system; including: The state anti-shake constraint condition of the UAV system in step 3 and the sufficient condition for the mean square stability of the UAV system in step 4 are converted into matrix inequalities. The matrix inequalities are combined to obtain the sufficient condition for the existence of a fault-tolerant controller with anti-shake performance. Step 6: Solve the matrix inequalities obtained in Step 5 to obtain the gain of the fault-tolerant controller of the UAV system; Step 7: Utilize the gain of the UAV system fault-tolerant controller to achieve active fault-tolerant control of the quadcopter UAV.

2. The active fault-tolerant control method for unmanned aerial vehicles with jitter suppression performance according to claim 1, characterized in that, The drone model described in step one is represented as follows: Where k represents time; Represents the state matrix of the unmanned aerial vehicle (UAV) system; Represents the control input matrix; , The positive constant s represents the sampling period of the system. Represents the identity matrix; and Represents the system matrix. This represents the uncertainty coefficient of the drone model. This represents the efficiency loss caused by fluctuations in the motor's output torque, and ; Let represent a stochastic process that follows a semi-Markov property, and this stochastic process is within a finite set. Take the value from; Representing the system matrix Matrix variables in Representing the system matrix Matrix variables in; This represents an uncertain parameter exhibiting multimodal characteristics.

3. The active fault-tolerant control method for unmanned aerial vehicles with jitter suppression performance according to claim 1, characterized in that, The expression for the piecewise homogeneous emission probability dominated by the Markov chain in step two is: in, Indicates the piecewise homogeneous time-varying emission probability; The observation mode representing FDI information; Represents the true modality of FDI information; Represents a high-dimensional Markov random chain; Indicates the length of stay; Representing the true modality of FDI information any value in, Observation modes representing FDI information any value in, Represents a high-dimensional Markov random chain The piecewise homogeneous time-varying emission probability is fixed within a time interval, and the switching between different time intervals is determined by a high-dimensional Markov chain. Dominate.

4. The active fault-tolerant control method for unmanned aerial vehicles with jitter suppression performance according to claim 3, characterized in that, The structure of the state stabilization controller described in step two is as follows: in, It is the controller gain to be solved.

5. The active fault-tolerant control method for unmanned aerial vehicles with jitter suppression performance according to claim 4, characterized in that, The lemma for determining the mean square stability of a system in step four is as follows: Consider discrete-time general switching systems with finite dwell times. The switching times are k0, k1, ..., kn, and k0 = 0; Represents a class of continuous-time functions; If for and ,in For a set of integers, From arrive For the set of integers that can take values, there exists a series of Lyapunov functions. and three function: , , For any initial conditions , , and given finite constants , , ,in The set of integers representing the random process to which the fault occurred. Represents the set of integers to which the FDI stochastic process belongs; We can obtain: in If the expected value is expressed, then the switching system is mean-square stable.

6. The active fault-tolerant control method for unmanned aerial vehicles with jitter suppression performance according to claim 5, characterized in that, In step four, based on the lemma for the mean square stability of a general switching system, the sufficient condition for the mean square stability of the UAV dynamic system in the absence of accurate FDI information is as follows: Considering the unmanned aerial vehicle control system and finite constants , , , If for , , , There exists a matrix set , And for , , The following inequalities are satisfied: The unmanned aerial vehicle control system It is mean-square stable; in, This represents the maximum dwell time of the system when the system mode is i, which is within the set of positive integers. The value is taken from h; h represents the set of integers. Any value in ; For intermediate process variables; , = .

7. The active fault-tolerant control method for unmanned aerial vehicles with jitter suppression performance according to claim 6, characterized in that, The sufficient condition for the existence of a fault-tolerant controller with anti-shake performance in step five is: Considering the unmanned aerial vehicle control system and finite constants , , , , finite matrix ; If for , , There exists a matrix set , , And for , , The following inequalities are satisfied: in, At this time, the drone control system It is mean-square stable and has a given anti-shake performance; wherein, the allowable controller gain is , Let be the set positive definite matrix.

Images