Buoy-unmanned aerial vehicle cooperative fault-tolerant docking control method based on event-triggered communication

Abstract

The invention discloses a buoy-unmanned aerial vehicle cooperative fault-tolerant docking control method based on event-triggered communication, and belongs to the field of unmanned aerial vehicle navigation, guidance and control. The method comprises the following steps: firstly, establishing a buoy-unmanned aerial vehicle affine nonlinear model considering the fault influence of an execution mechanism; then, a communication mechanism based on an event triggering strategy is designed, and in the collaborative docking process, aperiodic information interaction is carried out between the buoy and the unmanned aerial vehicle through the event triggering mechanism; then, based on a backstepping control theory, a controller is designed for the position, track, attitude and angular velocity loop of the buoy-unmanned aerial vehicle and the velocity loop of the unmanned aerial vehicle, and based on an observation-compensation idea, a lumped nonlinear item containing airflow disturbance and execution mechanism fault information is estimated and compensated; and finally, realizing cooperative fault-tolerant docking of the buoy and the unmanned aerial vehicle through the controller. According to the method, the accurate cooperative docking of the unmanned aerial vehicle and the buoy can still be realized under the conditions of limited communication resources, execution mechanism faults and complex airflow disturbance.

Description

Technical Field

[0001] The present invention belongs to the field of unmanned aerial vehicle (UAV) navigation, guidance and control, and in particular relates to a buoy-UAV collaborative fault-tolerant docking control method based on event-triggered communication. Background Art

[0002] Drone towed aerial recovery refers to the process of a small fixed-wing drone precisely docking with a buoy towed by a cable in the air, and then recovering the drone into the cargo hold of a transport aircraft by retracting the cable. Figure 1 As shown in the figure, drone aerial recovery technology eliminates the drone's reliance on reliable land-based or ship-based systems, effectively extending its flight range. During the recovery process, the drone's trajectory must be optimized to ensure smooth and rapid recovery. Precise docking and control of the drone and buoy directly determines the success of the recovery mission and is an urgent issue that needs to be addressed.

[0003] Unlike the drogue used in aerial refueling, which is typically uncontrollable, the buoy used in aerial recovery is equipped with grid rudders and can be actively controlled. Therefore, to further improve docking efficiency and accuracy while fully leveraging the buoy's controllable advantages, the docking mission can be accomplished through coordinated control of the buoy and drone. However, achieving coordinated docking control between the buoy and drone requires addressing the following key difficulties and challenges:

[0004] (1) The buoy-UAV docking system is a heterogeneous docking system. The buoy is an axisymmetric, unpowered, tethered and towed aircraft, while the UAV is a plane-symmetric, powered aircraft. The two have obvious differences in turning methods, aerodynamic characteristics, and drive devices. These differences increase the difficulty of control system design.

[0005] (2) Both the buoy and the drone towed by the flexible cable are subject to multiple complex airflow disturbances. In addition, complex environmental factors and long mission execution may cause actuator failures. These interferences and failures will greatly increase the difficulty of docking control and even threaten flight safety.

[0006] (3) Most existing collaborative control strategies require agents to exchange information within each control cycle. This continuous communication mechanism consumes a large amount of communication resources and places a heavy burden on the communication network.

[0007] Therefore, studying the cooperative fault-tolerant docking control of buoy-UAV heterogeneous systems under limited communication resources is of great significance for improving the docking efficiency and accuracy, as well as the safety of the aircraft. Summary of the Invention

[0008] To improve the docking accuracy and efficiency of drones and buoys, this paper proposes a buoy-drone collaborative fault-tolerant docking control method based on event-triggered communication. This method enables efficient and precise docking control between drones and buoys under adverse conditions such as complex airflow disturbances, actuator failures, and limited communication resources.

[0009] A buoy-UAV collaborative fault-tolerant docking control method based on event-triggered communication specifically includes the following steps:

[0010] Step 1: Establish a buoy-UAV affine nonlinear model considering the influence of actuator failure;

[0011] The specific steps are as follows:

[0012] Step 101: Establish an affine nonlinear model of the buoy and the UAV;

[0013] Collaborative docking system It consists of a towed buoy (marked as d) and a drone (marked as U), namely The 6-DOF affine nonlinear model of the buoy and the UAV is expressed as:

[0014]

[0015] in, X i1 =[y i ,z i ] T is the position loop state variable, X i2 =[χ i ,γ i ] T is the state variable of the track loop, X i3 =[α i ,β i ,μ i ] T is the attitude loop state variable, X i4 =[p i ,q i ,r i ] T is the state variable of the angular velocity loop, V Ug is the speed of the drone, υ i is an intermediate variable; F i1 、F i2 、F i3 、F i4 and is the lumped interference term of each loop, B i1 、B i2 、B i3 、B i4 and is the control input item of each loop; δ i =[δ ia ,δ ie ,δ ir ] T is the deflection angle of the aerodynamic control surface of the buoy and the UAV, δ UT is the throttle opening of the UAV. According to the model of the buoy and the UAV, the attitude loop state variable can be obtained by the intermediate variable υ i Calculated, they are X d3 =[υ d (1),υ d (2),μ d ] T and X U3 =[sign(υ U (2))||υ U ||,β U ,atan(υ U (1) / υ U (2))] T .

[0016] Step 102: Establish an actuator fault model;

[0017] Actuator failures are mainly considered to be loss of effectiveness and paranoid failures:

[0018]

[0019] in, ρ i =diag(ρ ia ,ρ ie ,ρ ir ) and ρ UT ∈(0,1] represents the unknown validity loss coefficient, 0<ρ ia ,ρ ie ,ρ ir ≤1; and for bounded paranoid failures; and They are the aerodynamic control surface deflection angle command and throttle opening command to be designed respectively.

[0020] Step 103: Establishing a buoy-UAV affine nonlinear model that takes into account the influence of actuator failure;

[0021] Combining the 6-DOF model of the aircraft and the above fault model, it can be seen that the actuator failure will directly affect the angular velocity motion equations of the buoy and the UAV, as well as the velocity motion equations of the UAV. Therefore, based on the above fault model, the angular velocity motion equations of the buoy and the UAV, as well as the velocity motion equations of the UAV, are modeled as follows:

[0022]

[0023] in, and Contains the fault information of the actuator, I 3×3 is a unit vector of size 3 × 3. The other loop motion equations of the buoy and the UAV remain unchanged.

[0024] Step 2: Design an event-triggered aperiodic communication mechanism;

[0025] During the collaborative docking process, the buoy and the drone send their own position information and receive each other's position information through the following event-triggered communication mechanism:

[0026]

[0027] Where, is the triggering moment of aircraft i, is the number of triggers; Indicates the sampling status, that is, the position information at the triggering moment; and are the parameters to be designed.

[0028] Only when the trigger conditions are met will aircraft i broadcast its current location information to the other party through the communication network Before the next event is triggered, aircraft i will no longer send its own information to the other party.

[0029] Step 3: Based on the backstepping control and anti-disturbance control theories, a buoy-UAV collaborative fault-tolerant docking controller is designed under the aperiodic communication mechanism to achieve collaborative fault-tolerant docking between the buoy and the UAV.

[0030] The specific steps are as follows:

[0031] Step 301: Design a position loop controller;

[0032] Taking into account the influence of virtual control law tracking error and filtering error, the dynamic equation of position loop tracking error is:

[0033]

[0034] Among them, e i1 =[e iy ,e iz ] T , F i1 =[F iy ,F iz ] T , B i1=diag(B iy ,B iz ), s i2 =[s iχ ,s iγ ] T is the filtering error,

[0035] In order to achieve the collaborative docking task, the relative position relationship between the buoy and the UAV and their respective tracking errors are comprehensively considered, and the collaborative docking errors of the buoy and the UAV are defined as follows:

[0036]

[0037] Among them, E i1 =[E iy ,E iz ] T , P Up is the position of the docking plug in the drone system; if a dU =1, it means the buoy can receive the information of the drone, otherwise, a dU =0;a Ud with a dU have the same meaning; b i =1 means that aircraft i can obtain the information of the expected docking position, otherwise, b i =0.

[0038] During the docking process, the docking plug on the back of the UAV is kept in the same position as the buoy. The center of mass position of the UAV in the collaborative error is converted into the position of the docking plug, that is, X U1 +R1R I / B P Up , R1=[0,1,0;0,0,1] is used to extract the lateral and longitudinal positions of the plug, R I / B is the transformation matrix from the UAV system to the inertial system. represents the conversion of the desired docking position to the desired center of mass position of the UAV.

[0039] Further considering the impact of event-triggered communication strategy, the triggered collaborative docking error in the aperiodic communication mode is defined as:

[0040]

[0041] in,

[0042] Taking the self-tracking error and the triggering coordination error as feedback items, the following coordinated docking virtual control law is designed:

[0043]

[0044] Among them, K i1 =diag(K iy ,K iz ) and k i is the feedback control gain; F i1 The estimated value of P i It is a parameter related to the communication topology. The specific solution steps are: is the Laplace matrix of the communication topology,

[0045] In order to obtain the command and its differential signal of the subsequent trajectory loop and avoid the "differential explosion" problem caused by repeated derivation in backstepping control, the virtual control command is filtered by the following first-order filtering link through dynamic surface control (DSC):

[0046]

[0047] Where, τ i2 is the time constant matrix.

[0048] Step 302: Design a track loop controller;

[0049] First, calculate the track loop tracking error e i2 The derivative of :

[0050]

[0051] Where,

[0052] is the stability error e i2 , the virtual control law of the designed trajectory loop is:

[0053]

[0054] Where K i2 =diag(K iχ ,K iγ ) is the feedback control gain, F i2 estimated value.

[0055] The attitude loop command and its differential signal are obtained through the following first-order inertia link:

[0056]

[0057] Where, is the time constant diagonal matrix.

[0058] According to the affine nonlinear model of the buoy and the UAV, the attitude angle instructions of the buoy and the UAV are calculated as follows:

[0059]

[0060] Step 303: Design an attitude loop controller;

[0061] Attitude loop tracking error e i3 Taking the derivative we can get:

[0062]

[0063] Based on this, the following virtual control law is designed:

[0064]

[0065] Where K i3 =diag(K iα ,K iβ ,K iμ ) is the feedback control gain, F i3 estimated value.

[0066] make Obtained through the following first-order filtering link and

[0067]

[0068] Where, τ i4 is the time constant matrix.

[0069] Step 304: Design an angular velocity loop controller;

[0070] Combined with the angular velocity motion equation considering the actuator failure, It can be calculated as:

[0071]

[0072] For the nonlinear term f containing the actuator fault information i4 The following fault-tolerant control law is designed based on the estimation and compensation of

[0073]

[0074] Where K i4 =diag(K ip ,K iq ,K ir ) is the feedback control gain, f i4 estimated value.

[0075] Step 305: Design the UAV speed loop controller;

[0076] A separate speed loop is designed for the UAV to change the relative forward distance between the UAV and the buoy. Combined with the definition of the UAV speed tracking error and the speed motion equation, Calculated as:

[0077]

[0078] Design the following fault-tolerant speed constraint controller:

[0079]

[0080] in, is the feedback gain.

[0081] The buoy-UAV collaborative fault-tolerant docking controller consists of a buoy control module and a UAV control module. Both modules include a position loop controller, a track loop controller, an attitude angle loop controller, and an angular velocity loop controller. The UAV control module also includes a velocity loop controller. For the buoy's position loop and the UAV's position and velocity loops, ATVBLF-based constraint controllers are designed, respectively, based on the physical requirements of the docking mission, to constrain the docking trajectory and velocity within a safe docking envelope. For the buoy's angular velocity loop and the UAV's angular velocity and velocity loops, EMLPNNs are used based on an "observe-and-compensate" principle to estimate and compensate for lumped nonlinear terms that include airflow disturbances and actuator fault information, thus ensuring the controllers' robustness and fault tolerance. During the collaborative docking process, intermittent information exchange occurs between the buoy and UAV via an event-triggered mechanism.

[0082] The advantages of the present invention are:

[0083] (1) A buoy-UAV collaborative fault-tolerant docking control method based on event-triggered communication improves docking efficiency and accuracy by collaboratively controlling the buoy and the UAV.

[0084] (2) A buoy-UAV collaborative fault-tolerant docking control method based on event-triggered communication. By designing a reasonable event-triggered communication mechanism, information exchange between the buoy and the UAV will only occur when the triggering conditions are met, rather than in every control cycle, thereby greatly reducing the communication frequency and effectively saving communication resources.

[0085] (3) A buoy-UAV collaborative fault-tolerant docking control method based on event-triggered communication considers the impact of actuator failure as lumped interference and compensates for it based on anti-disturbance control theory, thereby achieving precise docking under complex airflow disturbances and actuator failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 This is a schematic diagram of the buoy-UAV docking system;

[0087] Figure 2 This is a block diagram of the buoy-UAV heterogeneous system collaborative docking control of the present invention;

[0088] Figure 3 is a diagram of an actuator failure encountered by a buoy-UAV in an embodiment of the present invention;

[0089] Figure 4 is a three-dimensional trajectory diagram of the coordinated docking of the buoy and the UAV in an embodiment of the present invention;

[0090] Figure 5 1 is a diagram showing the wind disturbance conditions experienced by the buoy and the UAV in an embodiment of the present invention;

[0091] Figure 6 is a position tracking error diagram of a buoy in an embodiment of the present invention;

[0092] Figure 7 is a graph of position and velocity tracking errors of a UAV according to an embodiment of the present invention;

[0093] Figure 8 is a diagram showing the relationship between the sampling error and the trigger threshold in an embodiment of the present invention;

[0094] Figure 9 is a diagram of the triggering timing of the buoy and the UAV in an embodiment of the present invention;

[0095] Figure 10 This is a comparison chart of the number of communications under different communication mechanisms in an embodiment of the present invention;

[0096] Figure 11 This is a control input instruction diagram of a buoy in an embodiment of the present invention;

[0097] Figure 12 1 is a control input instruction diagram of the UAV in an embodiment of the present invention. DETAILED DESCRIPTION

[0098] In order to facilitate those skilled in the art to understand and implement the present invention, the present invention is further described in detail below with reference to the accompanying drawings and examples.

[0099] The present invention discloses a buoy-UAV collaborative fault-tolerant docking control method based on event-triggered communication. First, an affine nonlinear model of the buoy-UAV is established, which takes into account the influence of actuator failures. Then, a communication mechanism based on an event-triggered strategy is designed. During the collaborative docking process, the buoy and the UAV perform non-periodic information exchange through the event-triggered mechanism. Next, based on backstepping control theory, controllers are designed for the position, track, attitude, and angular velocity loops of the buoy-UAV, as well as the speed loop of the UAV. Based on the "observe-compensate" principle, lumped nonlinear terms containing airflow disturbances and actuator failure information are estimated and compensated, thereby making the controller have anti-disturbance and fault-tolerant capabilities.

[0100] A buoy-UAV collaborative fault-tolerant docking control method based on event-triggered communication, the specific implementation method includes the following steps:

[0101] Step 1: Establish a buoy-UAV affine nonlinear model considering the influence of actuator failure;

[0102] The specific steps are as follows:

[0103] Step 101: Establish an affine nonlinear model of the buoy and the UAV;

[0104] Collaborative docking system It consists of a towed buoy (marked as d) and a drone (marked as U), namely Referring to the modeling method of conventional aircraft, the 6-DOF (Degrees of Freedom) affine nonlinear model of buoys and UAVs can be expressed as:

[0105]

[0106] in, X i1 =[y i ,z i ] T is the position loop state variable, X i2 =[χ i ,γ i ] T is the state variable of the track loop, X i3 =[α i ,β i ,μ i ] T is the attitude loop state variable, X i4 =[p i ,q i ,r i ] T is the state variable of the angular velocity loop, V Ug is the speed of the drone, υi is an intermediate variable; F i1 、F i2 、F i3 、F i4 and is the lumped interference term of each loop, B i1 、B i2 、B i3 、B i4 and is the control input item of each loop, and the specific expression is not repeated here; δ i =[δ ia ,δ ie ,δ ir ] T is the deflection angle of the aerodynamic control surface of the buoy and the UAV, δ UT The throttle opening of the drone.

[0107] It's important to note that, unlike powered drones, buoys are unpowered, tethered, and towed objects. Their grid rudders can only control their longitudinal and lateral positions within a certain range, while their forward position is passively maintained at a relatively fixed position by the tension of the cables. Therefore, during the coordinated docking process, the docking plug and buoy are aligned laterally and longitudinally through coordinated control of the drone and buoy. In the forward passage, the relative forward distance between the drone and buoy is varied by controlling the drone's speed.

[0108] Step 102: Establish an actuator fault model;

[0109] The present invention mainly considers the following validity loss and paranoid fault models:

[0110]

[0111] in, ρ i =diag(ρ ia ,ρ ie ,ρ ir ) and ρ UT ∈(0,1] represents the unknown validity loss coefficient, 0<ρ ia ,ρ ie ,ρ ir ≤1; and for bounded paranoid failures; and are the aerodynamic control surface deflection angle command and throttle opening command to be designed respectively. i =diag(1,1,1),ρ UT =1 and δ i0 =[0,0,0] T , δUT0 =0, it means that the actuator has no fault; when 0<ρ ia ,ρ ie ,ρ ir <1、0<ρ UT <1 and δ UT0 =0, it means only the failure of loss of effectiveness occurs; when 0<ρ ia ,ρ ie ,ρ ir <1、0<ρ UT <1 and δ i0 ≠[0,0,0] T , δ UT0 When ≠0, it indicates that a mixed fault has occurred in the actuator.

[0112] Step 103: Establishing a buoy-UAV affine nonlinear model that takes into account the influence of actuator failure;

[0113] Combining the 6-DOF model of the aircraft and the above fault model, it can be seen that actuator failure will directly affect the angular velocity motion equations of the buoy and the UAV, as well as the velocity motion equations of the UAV. Therefore, based on the above fault model, the angular velocity motion equations of the buoy and the UAV, as well as the velocity motion equations of the UAV, can be modeled as follows:

[0114]

[0115] in, and The fault information of the actuator is included. The other loop motion equations of the buoy and the UAV remain unchanged.

[0116] Step 2: Design an event-triggered aperiodic communication mechanism;

[0117] During the collaborative docking process, the buoy and the drone send their own position information and receive each other's position information through the following event-triggered communication mechanism:

[0118]

[0119] Where, is the triggering moment of aircraft i, is the number of triggers; Indicates the sampling status, that is, the position information at the triggering moment; and are the parameters to be designed.

[0120] Only when the trigger conditions are met will aircraft i broadcast its current location information to the other party through the communication network Before the next event is triggered, aircraft i will no longer send its own information to the other party. This means that in the time interval For aircraft j, the available status information about the other aircraft i remains unchanged, that is, The sampling error can be defined as Through the above communication method, the aircraft only needs to send its own status information at the trigger moment, rather than transmitting information in every control cycle, which can greatly reduce the communication burden.

[0121] Unlike fixed-threshold event trigger strategies, the trigger threshold in the event trigger strategy designed by the present invention is described by a time exponential function, which has a monotonically decreasing characteristic. Generally speaking, although increasing the trigger threshold will reduce the communication frequency, it may sacrifice control accuracy, while lowering the trigger threshold will increase the communication frequency but may help improve control accuracy. In the transient phase, the state of the aircraft will undergo relatively large adjustments and changes, and the requirements for system control accuracy are relatively low, so the trigger threshold can be appropriately increased. On the contrary, in the steady-state phase, the trigger threshold should be lowered to achieve higher control accuracy. In addition, the trigger function designed by the present invention only depends on its own state information and does not depend on the state of other aircraft. It has the advantages of being simple in form and easy to implement.

[0122] Step 3: Based on backstepping control and anti-disturbance control theory, a buoy-UAV collaborative fault-tolerant docking controller is designed under the aperiodic communication mechanism.

[0123] The specific steps are as follows:

[0124] Step 301: Design a position loop controller;

[0125] Taking into account the influence of virtual control law tracking error and filtering error, the dynamic equation of position loop tracking error can be calculated as:

[0126]

[0127] Among them, e i1 =[e iy ,e iz ] T , F i1 =[F iy ,F iz ] T , B i1 =diag(B iy ,B iz ), s i2 =[s iχ ,s iγ ] T ,

[0128] In order to achieve the collaborative docking task, the relative position relationship between the buoy and the UAV and their respective tracking errors are comprehensively considered, and the collaborative docking errors of the buoy and the UAV are defined as follows:

[0129]

[0130] Among them, E i1 =[E iy ,E iz ] T , P Up is the position of the docking plug in the drone system; if a dU =1, it means the buoy can receive the information of the drone, otherwise, a dU =0; Similarly, if a Ud =1 means the UAV can receive the information from the buoy, otherwise, a Ud =0;b i =1 means that aircraft i can obtain the information of the expected docking position, otherwise, b i = 0. It should be noted that during the docking process, it is the docking plug on the back of the drone that needs to be consistent with the buoy position, not the center of mass of the drone. Therefore, the center of mass position of the drone in the collaborative error is X U1 Need to be converted to the position of the docking plug, that is, X U1 +R1R I / B P Up , R1=[0,1,0;0,0,1] is used to extract the lateral and longitudinal positions of the plug, R I / B is the transformation matrix from the UAV system to the inertial system. represents the conversion of the desired docking position to the desired center of mass position of the UAV.

[0131] Further considering the impact of event-triggered communication strategy, the triggered collaborative docking error in the aperiodic communication mode is defined as:

[0132]

[0133] in,

[0134] Taking the self-tracking error and the triggering coordination error as feedback items, the following coordinated docking virtual control law is designed:

[0135]

[0136] Among them, K i1 =diag(K iy ,K iz ) and k i is the feedback control gain; F i1 The estimated value of can be obtained by methods such as extended state observer or neural network; i It is a parameter related to the communication topology. The specific solution steps are: P = diag (P d ,P U )=diag(1 / q d ,1 / q U ), is the Laplace matrix of the communication topology,

[0137] In order to obtain the instructions and differential signals of the subsequent track loop and avoid the "differential explosion" problem caused by repeated derivation in backstepping control, the following first-order filtering link is used to filter the virtual control instructions:

[0138]

[0139] Where, τ i2 is the time constant matrix.

[0140] Step 302: Design a track loop controller;

[0141] First, calculate the track loop tracking error e i2 The derivative of :

[0142]

[0143] Where,

[0144] is the stability error e i2 , the virtual control law of the designed trajectory loop is:

[0145]

[0146] Where K i2 =diag(K iχ ,K iγ ) is the feedback control gain, F i2 estimated value.

[0147] The attitude loop command and its differential signal are obtained through the following first-order inertia link:

[0148]

[0149] Where, is the time constant diagonal matrix. According to the affine nonlinear model of the buoy and the UAV, the attitude angle instructions of the buoy and the UAV can be calculated as and

[0150] Step 303: Design an attitude loop controller;

[0151] Attitude loop tracking error e i3 Taking the derivative we can get:

[0152]

[0153] Based on this, the following virtual control law is designed:

[0154]

[0155] Where K i3 =diag(K iα ,K iβ ,K iμ ) is the feedback control gain, F i3 The estimated value of .

[0156] make Obtained through the following first-order filtering link and

[0157]

[0158] Where, τ i4 is the time constant matrix.

[0159] Step 304: Design an angular velocity loop controller;

[0160] Combined with the angular velocity motion equation considering the actuator failure, It can be calculated as:

[0161]

[0162] For the nonlinear term f containing the actuator fault information i4 The following fault-tolerant control law is designed based on the estimation and compensation of

[0163]

[0164] Where K i4 =diag(K ip ,K iq ,K ir ) is the feedback control gain, f i4 estimated value.

[0165] Step 305: Design the UAV speed loop controller;

[0166] For UAVs, a separate speed loop needs to be designed to change the relative forward distance between the UAV and the buoy. Combining the definition of UAV speed tracking error and the speed motion equation, It can be calculated as:

[0167]

[0168] Design the following fault-tolerant speed constraint controller:

[0169]

[0170] in, is the feedback gain.

[0171] After the above design, the buoy-UAV collaborative fault-tolerant docking control framework based on event-triggered communication mechanism is as follows: Figure 2 As shown, the system includes a buoy control module and a UAV control module. Both modules include a position loop controller, a track loop controller, an attitude angle loop controller, and an angular velocity loop controller. The UAV control module also includes a velocity loop controller. For the buoy's position loop and the UAV's position and velocity loops, ATVBLF-based constraint controllers are designed, respectively, based on the physical requirements of the docking mission, to constrain the docking trajectory and velocity within a safe docking envelope. For the buoy's angular velocity loop and the UAV's angular velocity and velocity loops, EMLPNNs are used based on an "observe-and-compensate" approach to estimate and compensate for lumped nonlinear terms that include airflow disturbances and actuator fault information, thus making the controllers both disturbance-resistant and fault-tolerant. During the collaborative docking process, intermittent information exchange occurs between the buoy and the UAV via an event-triggered mechanism.

[0172] Example

[0173] In order to verify the effectiveness and superiority of the present invention, a simulation verification is carried out by taking a certain type of UAV air recovery system as an example. The communication topology between the buoy and the UAV is set as a dU =a Ud =1,b d =b U = 1. The parameters of the event trigger mechanism are set to: a i =1,v i =-0.5, n i =0.05. The controller parameter is set to K Uy =K Uz =0.3, k U =0.1, K U2 =diag(2,4),K U3 =diag(6,1,2),KU4 =diag(20,10,10), K dy =K dz =0.5, k d =0.1, K d2 =diag(2,2),K d3 =diag(4,4,4),K d4 =diag(15,5,15). The initial positions of the drone and buoy are X U1 (0) = [15, 35] T m and X d1 (0) = [-5, 10] T m. The expected docking position is During the docking process, it is assumed that the actuator failure encountered by the buoy and the UAV is as follows: Figure 3 During the docking process, the buoy, towed by the transport aircraft, maintains a constant flight speed of 85 m / s; the drone needs to accelerate from 85 m / s to 87 m / s in 20 seconds to approach the buoy, and immediately decelerate to 85 m / s after successful docking to prevent the cable from slacking.

[0174] Figure 4 The three-dimensional collaborative docking trajectory of the buoy and the UAV docking plug is given. It can be seen that under the action of the controller, the positions of the buoy and the UAV plug eventually tend to the desired docking position, thus completing the docking task. Figure 5 The wind disturbances experienced by the buoy and the UAV during the coordinated docking process are given. The wake of the transport aircraft experienced by the buoy and the UAV changes with their respective positions. In addition, they are also affected by sudden gusts at 10 seconds.

[0175] Figure 6 The position tracking error of the buoy is given by Figure 7 The position and velocity tracking errors of the UAV are given. It is not difficult to see that under the action of the proposed collaborative controller, the tracking error can quickly converge to a smaller neighborhood, thereby improving the docking safety and success rate.

[0176] Figure 8 The relationship between the sampling error and the dynamic trigger threshold is shown. It can be seen that the trigger threshold decreases monotonically over time, especially in the first 10 seconds, the threshold decays rapidly, and then tends to be flat. When the sampling error exceeds the trigger threshold, it means that the current state of the aircraft has changed significantly compared to the state at the last trigger time. According to the event-triggered communication principle, the aircraft will send its current state information to the other party in a timely manner to ensure the reliability of the exchange information. Therefore, from Figure 8 It can be seen that whenever the sampling error exceeds the trigger threshold, the sampling error at the next moment will return to near 0.

[0177] Figure 9 The horizontal axis represents the time of the trigger moment, and the vertical axis represents the time interval between two adjacent trigger moments. Figure 8 and Figure 9 As can be seen, during the initial phase (the first 10 seconds), the states of the buoy and drone fluctuate significantly due to initial errors. Consequently, sampling errors easily exceed the trigger threshold, resulting in a high communication frequency. In the steady-state phase (after 10 seconds), both the buoy and drone have adjusted to the desired docking position. During this phase, their positions remain largely unchanged, sampling errors accumulate more slowly, and communication frequency is lower. Furthermore, compared to the buoy, the drone's initial error is greater, resulting in a greater rate and magnitude of position change during the initial phase. Consequently, the drone's communication frequency is higher than that of the buoy during the same period.

[0178] In order to better evaluate the advantages of the designed event trigger mechanism in reducing the communication frequency, Figure 10 The communication times under different communication strategies are compared. The trigger function in the fixed threshold trigger strategy is set to Compared with time-triggered and fixed-threshold event-triggered communication mechanisms, the event-triggered communication mechanism designed in the present invention can significantly reduce the communication frequency and save communication resources.

[0179] Figure 11 and Figure 12 The control input instructions for the buoy and the UAV are given respectively, combined with Figure 3 It can be seen that after the actuator fails, the control law can adjust the control input signal in time to compensate for the impact of the actuator failure, thereby maintaining the control accuracy.

[0180] The above simulation verification of the embodiments proves the effectiveness of the buoy-UAV collaborative fault-tolerant docking control method based on event-triggered communication of the present invention.

[0181] The contents not described in detail in the specification of the present invention belong to the prior art known to those skilled in the art.

Claims

1. A buoy-UAV collaborative fault-tolerant docking control method based on event-triggered communication, characterized in that: The specific steps include: Step 1: Establish a buoy-UAV affine nonlinear model considering the influence of actuator failure; Based on the affine nonlinear models of the buoy and UAV and the actuator failure model, a buoy-UAV affine nonlinear model considering the influence of actuator failure is established, which is expressed as: in, Collaborative docking system d is a towed buoy, U is a drone; X i1 =[y i ,z i ] T is the position loop state variable, X i2 =[χ i ,γ i ] T is the state variable of the track loop, X i3 =[α i ,β i ,μ i ] T is the attitude loop state variable, X i4 =[p i ,q i ,r i ] T is the state variable of the angular velocity loop, V Ug is the speed of the drone, υ i is an intermediate variable; and Contains the fault information of the actuator, I 3×3 is a unit vector of size 3×3, ρ i =diag(ρ ia ,ρ ie ,ρ ir ) and ρ UT ∈(0,1] represents the unknown validity loss coefficient, 0<ρ ia ,ρ ie ,ρ ir ≤1; and for bounded paranoid failures; and are the aerodynamic control surface deflection angle command and throttle opening command to be designed respectively; F i1 、F i2 、F i3 、F i4 and is the lumped interference term of each loop, B i1 、B i2 、B i3 、B i4 and It is the control input item of each loop; Step 2: Design an event-triggered aperiodic communication mechanism; During the collaborative docking process, the buoy and the drone send their own position information and receive each other's position information through the following event-triggered communication mechanism: in, is the triggering moment of aircraft i, is the number of triggers; Indicates the sampling status, that is, the position information at the triggering moment; and are the parameters to be designed; When the trigger condition is met, aircraft i broadcasts its current location information to the other party through the communication network Before the next event triggering moment, aircraft i will no longer send its own information to the other party; Step 3: Based on the backstepping control and anti-disturbance control theories, a buoy-UAV collaborative fault-tolerant docking controller is designed under the aperiodic communication mechanism to achieve collaborative fault-tolerant docking between the buoy and the UAV. The specific steps are as follows: Step 301: Design a position loop controller; Taking into account the influence of virtual control law tracking error and filtering error, the dynamic equation of position loop tracking error is: Among them, e i1 =[e iy ,e iz ] T , F i1 =[F iy ,F iz ] T , B i1 =diag(B iy ,B iz ), s i2 =[s iχ ,s iγ ] T is the filtering error, Taking the tracking errors of the buoy and the UAV and the error in triggering the coordinated docking as feedback items, the following coordinated docking virtual control law is designed: Among them, K i1 =diag(K iy ,K iz ) and k i is the feedback control gain; F i1 The estimated value of P i It is a parameter related to the communication topology; The virtual control instructions are filtered using the following first-order filtering link: Among them, τ i2 is the time constant matrix; Step 302: Design a track loop controller; First, calculate the track loop tracking error e i2 The derivative of : in, is the stability error e i2 , the virtual control law of the designed trajectory loop is: Among them, K i2 =diag(K iχ ,K iγ ) is the feedback control gain, F i2 estimated value of; The attitude loop command and its differential signal are obtained through the following first-order inertia link: in, is the time constant diagonal matrix; According to the affine nonlinear model of the buoy and the UAV, the attitude angle instructions of the buoy and the UAV are calculated as follows: Step 303: Design an attitude loop controller; Attitude loop tracking error e i3 Taking the derivative we can get: Based on this, the following virtual control law is designed: Among them, K i3 =diag(K iα ,K iβ ,K iμ ) is the feedback control gain, F i3 estimated value of; make Obtained through the following first-order filtering link and Among them, τ i4 is the time constant matrix; Step 304: Design an angular velocity loop controller; Combined with the angular velocity motion equation considering the actuator failure, Calculated as: For the nonlinear term f containing the actuator fault information i4 The following fault-tolerant control law is designed based on the estimation and compensation of Among them, K i4 =diag(K ip ,K iq ,K ir ) is the feedback control gain, f i4 estimated value of; Step 305: Design the UAV speed loop controller; A separate speed loop is designed for the UAV to change the relative forward distance between the UAV and the buoy. Combined with the definition of the UAV speed tracking error and the speed motion equation, Calculated as: Design the following fault-tolerant speed constraint controller: in, is the feedback gain; Step 306, the buoy-UAV collaborative fault-tolerant docking controller includes a buoy control module and a UAV control module. The buoy control module and the UAV control module both include a position loop controller, a track loop controller, an attitude angle loop controller and an angular velocity loop controller. The UAV control module also includes a velocity loop controller.

2. The method for controlling buoy-UAV collaborative fault-tolerant docking based on event-triggered communication according to claim 1, characterized in that: The affine nonlinear model of the buoy and the UAV is expressed as: Among them, δ i =[δ ia ,δ ie ,δ ir ] T is the deflection angle of the aerodynamic control surface of the buoy and the UAV, δ UT The throttle opening of the drone.

3. The method for controlling buoy-UAV collaborative fault-tolerant docking based on event-triggered communication according to claim 1, characterized in that: The actuator failure is a loss of effectiveness and paranoid failure model: The actuator failure has a direct impact on the angular velocity motion equations of the buoy and the UAV, as well as the velocity motion equations of the UAV. The forms of the other loop motion equations of the buoy and the UAV remain unchanged.

4. The method for controlling buoy-UAV collaborative fault-tolerant docking based on event-triggered communication according to claim 1, characterized in that: The self-tracking error of the buoy and the UAV is expressed as: Among them, E i1 =[E iy ,E iz ] T , P Up is the position of the docking plug in the drone system; if a dU =1, it means the buoy can receive the information of the drone, otherwise, a dU =0;a Ud with a dU have the same meaning; b i =1 means that aircraft i can obtain the information of the expected docking position, otherwise, b i =0; During the docking process, the docking plug on the back of the UAV is kept in the same position as the buoy. The center of mass position of the UAV in the collaborative error is converted into the position of the docking plug, that is, X U1 +R1R I / B P Up , R1=[0,1,0;0,0,1] is used to extract the lateral and longitudinal positions of the plug, R I / B is the transformation matrix from the UAV system to the inertial system; represents the conversion of the desired docking position to the desired center of mass position of the UAV.

5. The method for controlling buoy-UAV collaborative fault-tolerant docking based on event-triggered communication according to claim 1 or 4, characterized in that: The error of triggering the coordinated docking between the buoy and the UAV is expressed as: in, 6. The method for controlling buoy-UAV collaborative fault-tolerant docking based on event-triggered communication according to claim 1, characterized in that: The application process of the buoy-UAV collaborative fault-tolerant docking controller is as follows: For the buoy's position loop and the UAV's position and velocity loops, constraint controllers based on ATVBlF were designed, respectively, in response to the physical requirements of the docking mission. This approach constrains the docking trajectory and velocity within a safe docking envelope. For the buoy's angular velocity loop and the UAV's angular velocity and velocity loops, EMLPNNs were used based on an "observe-and-compensate" approach to estimate and compensate for lumped nonlinear terms that include airflow disturbances and actuator fault information, thus ensuring the controllers' robustness and fault tolerance. During the collaborative docking process, intermittent information exchange between the buoy and UAV was carried out through an event-triggered mechanism.

Images