Cluster unmanned aerial vehicle cooperative toughness control method for DoS strike

Abstract

The invention discloses a cluster unmanned aerial vehicle cooperative toughness control method for a link DoS attack, and relates to the technical field of unmanned aerial vehicles, and the method comprises the steps: building a virtual adjacent vehicle based on the latest information received by a follower unmanned aerial vehicle before the follower unmanned aerial vehicle encounters the link DoS network attack; the designed virtual neighbor machine meets the requirement that the derivative of the positive direction position reference state, the position state of the side direction and the position state of the rising direction are kept unchanged before encountering the link DoS attack, and a communication topology passive switching mechanism under the link DoS network attack is designed based on the state of the virtual neighbor machine. And then a fault parameter estimation-based tough fault-tolerant cooperative controller is designed, so that the follower unmanned aerial vehicle can realize passive link switching under the condition that the follower unmanned aerial vehicle encounters a link DoS network attack and cannot receive neighbor information, and meanwhile, parameter estimation and fault-tolerant control on an actuator fault can be realized. The method has good practical significance and application prospects in the field of formation cooperative fault-tolerant toughness control of multiple unmanned aerial vehicles.

Description

Technical Field

[0001] The present application relates to the field of UAV technology, and in particular to a resilient fault-tolerant collaborative control method for clustered UAVs against link DoS network attacks and actuator failures. Background Art

[0002] With the rapid development of drone technology, swarm drone systems have been extended to diversified fields such as agricultural precision operations, urban air traffic management, and disaster relief. In order to achieve collaborative operations of swarm drones in multiple fields and ensure the stability and safety of mission execution, distributed collaborative control strategies have become the core focus and frontier exploration direction of drone technology development.

[0003] Under the distributed cooperative control architecture of swarm drones, drones rely on communication networks for internal information exchange, so that each drone can obtain the status data of neighboring drones in real time, so as to perform tasks in a coordinated and consistent manner. However, this cooperative mechanism that relies on communication networks also faces severe external security challenges, especially DoS (Denial of Service) network attacks. Such attacks aim to cut off or interfere with the normal communication links between drones, resulting in obstruction of information transmission, making it impossible for drones to obtain the status of neighboring drones in a timely manner, destroying the coordination of the entire cluster, and posing a major threat to the security and stability of the cluster. In addition, as a key component for executing flight instructions, the actuator is responsible for driving various actions of the drone. Therefore, drone actuator failure is a potential and serious risk in flight missions. When an actuator fails, it may not be able to accurately execute the instructions of the flight control system, causing the drone to deviate from the predetermined route, unstable flight attitude, and even cause a crash. Summary of the invention

[0004] In response to the above problems and technical requirements, this application proposes a resilient fault-tolerant collaborative control method for UAVs against link DoS network attacks and actuator failures.

[0005] The technical solution of this application is as follows:

[0006] A cluster UAV collaborative resilience control method for DoS attacks, the control method comprising the following steps:

[0007] S1 builds a virtual neighbor based on the information received by the follower drone before encountering the link DoS network attack;

[0008] S2 designs a communication topology passive switching mechanism under a link DoS network attack based on the state of the virtual neighbor machine to determine a specific value of the communication link weight;

[0009] S3 is based on the passive topology communication switching mechanism and combines the UAV's dynamic model to obtain a collaborative resilient fault-tolerant collaborative controller for follower UAVs under link DoS network attacks and actuator failures.

[0010] Furthermore, the S1 is specifically:

[0011] S11 assumes that the swarm drone system includes a real or virtual leader drone and N real follower drones, where any follower drone is i, 1≤i≤N:

[0012] Define the full state tracking deviation of follower UAV i

[0013] Among them, X 1，i is the actual position of follower drone i, X 1，j represents the actual position of the jth neighbor aircraft, X 1，j represents the actual position of the jth neighbor aircraft, D i，j represents the relative position between the i-th UAV and the j-th UAV;

[0014] Indicates the position of the jth neighboring aircraft when encountering a DoS attack, Y 1，i Indicates the rolling angle, angle of attack and sideslip angle Y of the drone 1，i =[μ i , α i , β i ],Y 2，i Indicates the three-axis angular velocity Y of the drone 2，i =[p i ,q i , r i ], a i，j represents the weight of the inter-machine communication link;

[0015] Indicates the communication link weight under the link switching mechanism when encountering a DoS attack; is a virtual control signal;

[0016] S12 Design of Control Deviation for UAV Cooperative Control c3, c6, and c7 are the positive parameters of the controller to be designed.

[0017] Furthermore, the S2 is specifically:

[0018] S21 uses the neighbor information obtained by follower UAV i through the cluster UAV communication network when it is not subjected to the link DoS network attack as the virtual neighbor information during the link DoS network attack;

[0019] S22 constructs the following intermediate variables:

[0020]

[0021]

[0022]

[0023] in, Represents the communication link weight a i，j No link DoS network attack occurred at time t; represents the latest updated position status of the jth UAV; τ0 represents the system sampling period;

[0024] S23 Design communication link weight a i，j Passive switching mechanism to obtain communication link weight for:

[0025]

[0026] Among them, a i，j is the actual communication link weight under the link DoS network attack. i，j When a DoS network attack occurs, i，j will be forced by a i，j =1 changes to a i，j =0, based on the actual communication link weight a i，j .

[0027] Furthermore, the S3 is specifically:

[0028] S31 establishes the dynamic model of follower UAV i as:

[0029]

[0030]

[0031]

[0032]

[0033]

[0034] where x i ,y i , z i represents the position of the i-th UAV, V i , χ i , γ i represents the UAV speed, track angle and ground track angle, μi , α i , β i is the roll angle, attack angle and sideslip angle. i ,q i , r i represents the angular velocity of the i-th UAV, is the aerodynamic moment of the i-th UAV;

[0035]

[0036]

[0037] Among them, δ ia , δ ie , δ ir are the aileron, elevator and rudder deflections of the i-th UAV respectively; T i , D i , L i , Y i Indicates thrust, drag, lift and lateral force; Q i ,s i , b i , c i Express dynamic pressure, wing area, wingspan, and mean aerodynamic chord;

[0038] in, c i2 =(I xi -I yi +I zi )I xzi / ∑ i , c i3 =I zi / ∑ i , c i4 =I xzi / ∑ i , c i5 =(I zi -I xi ) / I yi , c i6 =I xzi / I yi , c i7 =1 / I yi , c i9 =I xi / ∑ i , I xz , I yi , I zi represents the moment of inertia, I xzi is the moment of inertia, C iD0 , C iDα , CiDα2 , C iY0 , C iYβ , C iL0 , C iLα , C il0 , C ilβ , C ilp , C ilr , C im0 , C imα , i mq , C in0 , C inβ , C inp , and C inr represents the relevant aerodynamic parameters;

[0039] S32 then converts the dynamic model of the follower UAV i to obtain:

[0040]

[0041]

[0042] Among them, f 1，i , f 2，i , f 3，i , f 4，i , g 1，i , g 2，i , g 3，i , g 4，i represents the corresponding control vector and control matrix, which are derived from the dynamic model of UAV i; X 1，i =[x i ,y i ,h i ] T , X 2，i =[ V i, χ i , γ i ] T , i =[T i , α i sinμ i , α i cosμ i ] T , Y 1，i =[μ i , α i , β i ] T , Y 2，i =[p i ,q i , r i ]T ,u i =[δ ia , δ ie , δ ir ] T ; At the same time, there are:

[0043]

[0044] S33 tracking error e i，1 Taking the second derivative yields:

[0045]

[0046] in, By taking the derivative of the dynamic model we get:

[0047]

[0048] Design a sliding surface of the following form:

[0049]

[0050] Sliding surface S i，1 The derivative is:

[0051]

[0052] S34 According to S i，1 Expression and control deviation for UAV cooperative control defined in this application Then design the following virtual control signal:

[0053]

[0054] S35 according to the virtual control signal The resilient fault-tolerant collaborative controller of follower UAV i considering link DoS network attack and actuator failure is obtained as:

[0055]

[0056] Furthermore, according to the virtual control signal Design the intermediate virtual control signal of the follower drone as:

[0057]

[0058] Furthermore, the resilient fault-tolerant collaborative controller also includes an actuator fault parameter and Designed to estimate and The adaptive law of:

[0059]

[0060]

[0061] The defined variables satisfy the following relationship:

[0062]

[0063]

[0064] The beneficial technical effects of this application are:

[0065] The present application discloses a method for resilient fault-tolerant collaborative control of cluster UAVs against link DoS network attacks and actuator failures, so that the follower UAV can achieve passive link switching when encountering a link DoS network attack and unable to obtain neighboring machine information, thereby ensuring the stability of the cluster UAVs when encountering a link DoS network attack; at the same time, parameter estimation and fault-tolerant control of actuator failures can be achieved, which has great practical significance and application prospects in the formation collaborative fault-tolerant resilient control of multiple UAVs. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a topological diagram of the communication network of swarm drones in an example;

[0067] Figure 2 This is a schematic diagram of the information flow of a resilient fault-tolerant collaborative controller constructed in one embodiment of the present application;

[0068] Figure 3 It is a control block diagram of a cluster UAV resilient fault-tolerant collaborative control method in one embodiment of the present application;

[0069] Figure 4 is a timing diagram of the network attack signals imposed on each follower drone in a simulation instance;

[0070] Figure 5 yes Figure 4 The flight trajectory simulation diagram of each follower UAV in the simulation example;

[0071] Figure 6 yes Figure 4 The fault parameter estimation result diagram of each follower UAV in the simulation example;

[0072] Figure 7 yes Figure 4 Simulation results of the tracking error of each follower UAV in the simulation example. DETAILED DESCRIPTION

[0073] The specific implementation of the present application is further described below in conjunction with the accompanying drawings.

[0074] This application discloses a method for resilient fault-tolerant collaborative control of swarm drones against link DoS network attacks and actuator failures. In a swarm drone system, a communication network is built between the drones, and the network constructs a communication topology framework with the leader drone as the center. Under this architecture, the leader drone serves as the central node and establishes a one-way communication link with some follower drones, and the leader transmits location information to the followers. The follower drones can implement two-way or one-way communication interaction based on actual needs. For example, Figure 1 In the example, the cluster UAVs include 1 leader UAV (denoted as UAV#Leader) and 3 follower UAVs, denoted as UAV#1, UAV#2, and UAV#3 respectively. UAV#Leader establishes unidirectional communication connections with UAV#1 and UAV#2 respectively, UAV#1 establishes a unidirectional communication connection with UAV#2, UAV#2 establishes a unidirectional communication connection with UAV#3, and UAV#1 establishes a bidirectional communication connection with UAV#3.

[0075] Please refer to Figure 2 The information flow diagram shown and Figure 3 As shown in the control block diagram, the cluster UAV resilient fault-tolerant collaborative control method includes performing the following operations for any follower UAV i in the cluster UAV, with integer parameters 1≤i≤N:

[0076] Tracking deviation according to the state of follower UAV i For the topology of the swarm UAV communication network, Figure 1 In the example of 0，1 =a 0，2 =a 1，2 =a 1，3 =a 2，3 =a 3，1 =1, the rest a i，j = 0. For any follower drone i in the swarm of drones, its actual position X during the flight 1，i =[x i ,y i ,h i ], x i ,y i and h i They represent the actual position of the i-th follower UAV on the three coordinate axes of the ground coordinate system. 1，j D is the actual position of other drone j obtained through the cluster drone communication network. i，j is the expected relative position between follower UAV i and other UAVs j and is a known quantity.

[0077] Considering the link DoS network attack, and when the communication link weight a i，j When encountering a DoS network attack, the communication link resources between follower drone i and follower drone j will be occupied by the intruder, and follower drone i will not be able to obtain the status information of drone j through the cluster drone communication network, nor will it be able to obtain the location reference status through the cluster drone communication network according to the above traditional method.

[0078] Therefore, this application is based on the communication link weight a i，j The UAV state when not encountering a link DoS network attack, the design meets the following characteristics of the virtual follower UAV j flight requirements when encountering a DoS network attack, and designs a passive link switching mechanism under DoS attack to ensure the safe flight of the UAV when encountering a link DoS network attack.

[0079] Communication link weight a i，j The flight requirements that the virtual follower UAV j meets when encountering a link DoS network attack include: (1) When encountering a link DoS network attack, the state of the virtual follower UAV j will maintain the current forward direction of flight. At this time, the derivative of the position reference state of the follower UAV j in the positive direction remains unchanged before the link DoS network attack. (2) When encountering a link DoS network attack, the state of the virtual follower UAV j does not adopt yaw and lift flight. That is, the derivative of the position reference state of the follower UAV i in the lateral direction and lift is 0.

[0080] Based on the above characteristics, the following intermediate variables are first constructed:

[0081]

[0082]

[0083]

[0084] in, Represents the communication link weight a i，j There is no link DoS network attack at time t. represents the latest updated position status of the jth UAV. τ0 represents the system sampling period. Furthermore, the communication link weight a is designed i，j Passive switching mechanism to obtain communication link weight for:

[0085]

[0086] Among them, a i，j is the actual communication link weight under the link DoS network attack.i，j When a DoS network attack occurs, i，j will be forced by a i，j =1 changes to a i，j =0, based on the actual communication link weight a i，j ,This patent will design a passive topology communication switching mechanism to ,achieve smooth switching of the communication link. ,In addition, c1, c2, p, q are parameters to be designed.

[0087] Then, based on the above passive topology communication switching mechanism and combined with the dynamic model of UAV i, the collaborative resilience fault-tolerant collaborative controller of follower UAV i considering link DoS network attack and actuator failure is obtained, which includes the following steps:

[0088] (1) First, the dynamic model of follower UAV i is established as:

[0089]

[0090]

[0091]

[0092]

[0093]

[0094] where x i ,y i , z i V represents the position of the i-th UAV. i , χ i , γ i Represents the UAV speed, track angle and ground track angle. i , α i , β i is the roll angle, attack angle and sideslip angle. i ,q i , r i represents the angular velocity of the i-th UAV. is the aerodynamic torque of the i-th UAV.

[0095]

[0096]

[0097] Among them, δ ia , δ ie , δ ir are the aileron, elevator and rudder deflections of the i-th UAV respectively. i , D i , L i , Yi It represents thrust, drag, lift and lateral force. i ,s i , b i , c i represents the dynamic pressure, wing area, wingspan and mean aerodynamic chord, ρ represents the air density. Among them, c i2 =(I xi -I yi +I zi )I xzi / ∑ i , c i3 =I zi / ∑ i , c i4 =I xzi / ∑ i , c i5 =(I zi -I xi ) / I yi , c i6 =I xzi / I yi , c i7 =1 / I yi , c i9 =I xi / ∑ i , I xi , I yi , I zi represents the moment of inertia, I xzi is the moment of inertia. iD0 , C iDα , C iDα2 , C iY0 , C iYβ , C iL0 , C iLa , C il0 , C ilβ , C ilp , C ilr , C im0 , C imα , C imq , C in0 , C inβ , C inp , and C inr Represents relevant aerodynamic parameters.

[0098] (2) Then the dynamic model of follower UAV i is converted to:

[0099]

[0100]

[0101] Among them, f 1，i , f 2，i , f 3，i , f 4，i , g 1，i , g 2，i , g 3，i , g 4，i represents the corresponding control vector and control matrix, which are derived from the dynamic model of UAV i. 1，i =[x i ,y i ,h i ] T , X 2，i =[V i , χ i , γ i ] T , i =[T i , α i sinμ i , α i cosμ i ] T , Y 1，i =[μ i , α i , β i ] T , Y 2，i =[p i ,q i , r i ] T ,u i =[δ ia , δ ie , δ ir ] T . At the same time, there are:

[0102]

[0103] (3) Tracking error e i，1 Taking the second derivative yields:

[0104]

[0105] in, By taking the derivative of the kinetic model in (2), we can obtain:

[0106]

[0107] According to the control deviation for UAV cooperative control defined in this application Design a sliding surface of the following form:

[0108]

[0109] Sliding surface S i，1 The derivative is:

[0110]

[0111] (4) According to S i，1 Expression and control deviation for UAV cooperative control defined in this application Then design the following virtual control signal:

[0112]

[0113] Furthermore, according to the virtual control signal Design the intermediate virtual control signal of the follower drone as:

[0114]

[0115] (5) According to the virtual control signal The resilient fault-tolerant collaborative controller of follower UAV i considering link DoS network attack and actuator failure is obtained as:

[0116]

[0117] Among them, u f，i Represents the controller input u after filter processing i Based on the designed resilient fault-tolerant cooperative controller, it can be seen that the actuator fault parameters are also included. and Therefore, they are also designed to estimate and The adaptive law of:

[0118]

[0119]

[0120] The defined variables satisfy the following relationship:

[0121]

[0122]

[0123] Once the design is completed, the communication link weight can be designed using the passive link switching mechanism during the flight of the swarm drones. Realize the communication topology switching under link DoS attack, use the virtual neighbor information to determine the reference state of the UAV during the link attack, and use the UAV flight state to design the actuator fault parameter observer and Estimate the fault parameter ρ i and Then, according to the resilient fault-tolerant collaborative controller, the UAV control input u is obtained. i , and sends the control input signal to the follower UAV i for control. Thus, the resilient fault-tolerant collaborative controller is used to perform resilient fault-tolerant collaborative control on the follower UAV i to alleviate the impact of link DoS attacks and actuator failures on the cluster collaborative characteristics, and improve the security of the overall cluster system.

[0124] In order to verify the effectiveness of this method, based on Figure 1 The topology of the cluster UAV communication network shown in the figure constructs a simulation example. Figure 1 In the figure, the leader drone is numbered V0, and the follower drones are numbered V1, V2, and V3. The directed arrows indicate the direction of information flow between machines, and the dotted lines indicate the communication links with link DoS attacks. Follower drones V1, V2, and V3 encounter actuator failures in the simulation instance.

[0125] The values ​​of the structural parameters and aerodynamic coefficients of each UAV are as follows:

[0126] The acceleration due to gravity is g a =9.8kg / m 2 , the mass of all drones is m i =25kg, air density is ρ=1.2205kg / m 3 , the wing area of ​​all drones is s i =1.463m 2 , the wingspan is b i =2.8956, the average aerodynamic chord is c i =0.451, the moment of inertia is I xi =9.416,I yi =6.765,I zi =16.08, the moment of inertia is I xzi =0.1685. The aerodynamic parameter is C iD0 =0.0225, C iDα =0.1002, C iDα2 =1.0778, C iY，0 =0, C iYβ =-0.0046, C iL0 =0.2153, C iLα =4.6333, C il0 =0, Cilβ =-0.00054, C ilp =-0.49934, C ilr =0.07246, C im0 =-0.09543, C imα =-2.8206, C imq =-13.81889, C in0 =0, C inβ =0.00115, C inp =-0.03313, C inr =-0.04552.

[0127] The control parameters in the resilient fault-tolerant collaborative control are designed as follows:

[0128] p=1.4, q=0.6, c1=0.6, c2=0.4, c3=[10, 15, 15] T , c4=diag([3,1.8,0.8]), c5=diag([10 -4 , 10 -4 , 10 -4 ]), c6 = [1, 1, 1] T , c7 = [1, 1, 1] T , c8=0.6, K1=diag([8,1,2.3]), K2=diag([5,5,5]), K3=diag([10,10,10]), K4=diag([10,10,10]), K5=diag([10,10,10]).

[0129] The initial motion parameters of the swarm drone are set as follows:

[0130] The motion trajectory of the virtual leader UAV is used as the reference trajectory of the cluster UAV. The reference trajectory is X 1，0 =[20t+20cos(0.05t+0.1),0,0] T , the initial state of all follower drones is set to: X 1，1 (0) = [5, 5, 1] T , X 1，2 (0) = [5, 5, 1] T , X 1，3 (0) = [6, 2, -8] T , X 2，1 (0) = [19, -0.4, 0.2] T , X 2，2 (0) = [16, 0.15, -0.1] T, X 2，3 (0) = [21, -0.1, -0.1] T , Y 1，1 (0) = [1.15, 0.25, -0.05] T , Y 1，2 (0) = [0.8, -0.15, 0.1] T , Y 1，3 (0) = [-0.2, 0.4, 0.1] T , Y 2，1 (0) = Y 2，2 (0) = Y 2，3 (0) = [0, 0, 0] T .

[0131] The expected relative position between each follower drone and the leader drone is set to (in meters):

[0132] D 1，0 =[0,-5sin(0.2t),-5cos(0.2t)] T , D 2，0 =[0,-5sin(0.2t+2π / 3),-5cos(0.2t+2π / 3)] T , D 3，0 =[0,-5sin(0.2t+4π / 3),-5cos(0.2t+4π / 3)] T .

[0133] The actuator fault applied is:

[0134]

[0135] The flight trajectory of each follower drone in the cluster of drones obtained by simulation is as follows: Figure 5 The actuator fault parameter results of each follower UAV are shown in Figure 6 As shown. The tracking error e of each follower UAV i =[e i1 ,e i2 ,e i3 ] T The simulation curve is as follows Figure 7 shown.

[0136] The above is only a preferred embodiment of the present application, and the present application is not limited to the above embodiments. It is understood that other improvements and changes directly derived or associated by those skilled in the art without departing from the spirit and concept of the present application should be considered to be included in the protection scope of the present application.

Claims

1. A cluster UAV collaborative resilience control method for DoS attacks, characterized in that: The control method comprises the following steps: S1 builds a virtual neighbor based on the information received by the follower drone before encountering the link DoS network attack; S2 designs a communication topology passive switching mechanism under a link DoS network attack based on the state of the virtual neighbor machine to determine a specific value of the communication link weight; S3 is based on the passive topology communication switching mechanism and combines the UAV's dynamic model to obtain a collaborative resilient fault-tolerant collaborative controller for follower UAVs under link DoS network attacks and actuator failures.

2. The method for controlling the coordinated resilience of clustered drones against DoS attacks according to claim 1 is characterized in that: The S1 is specifically: S11 assumes that the swarm drone system includes a real or virtual leader drone and N real follower drones, where any follower drone is i, 1≤i≤N: Define the full state tracking deviation of follower UAV i Among them, X 1,i is the actual position of follower drone i, X 1,j represents the actual position of the jth neighbor aircraft, X 1,j represents the actual position of the jth neighbor aircraft, D i,j represents the relative position between the i-th UAV and the j-th UAV; Indicates the position of the jth neighboring aircraft when encountering a DoS attack, Y 1,i Indicates the rolling angle, angle of attack and sideslip angle Y of the drone 1,i =[μ i ,α i ,β i ],Y 2,i Indicates the three-axis angular velocity Y of the drone 2,i =[p i ,q i ,r i ], a i,j represents the weight of the inter-machine communication link; Indicates the communication link weight under the link switching mechanism when encountering a DoS attack; is a virtual control signal; S12 Design of Control Deviation for UAV Cooperative Control c3, c6, and c7 are the positive parameters of the controller to be designed.

3. The method for controlling the coordinated resilience of clustered drones against DoS attacks according to claim 1 is characterized in that: The S2 is specifically: S21 uses the neighbor information obtained by follower UAV i through the cluster UAV communication network when it is not subjected to the link DoS network attack as the virtual neighbor information during the link DoS network attack; S22 constructs the following intermediate variables: in, represents the communication link weight a i,j No link DoS network attack occurred at time t; represents the latest updated position status of the jth UAV; τ0 represents the system sampling period; S23 Design communication link weight a i,j Passive switching mechanism to obtain communication link weight for: Among them, a i,j is the actual communication link weight under the link DoS network attack. i,j When a DoS network attack occurs, i,j will be forced by a i,j =1 changes to a i,j =0, based on the actual communication link weight a i,j .

4. The method for controlling the coordinated resilience of clustered drones against DoS attacks according to claim 1 is characterized in that: The S3 is specifically: S31 establishes the dynamic model of follower UAV i as: where x i ,y i , z i represents the position of the i-th UAV, V i , χ i , γ i represents the UAV speed, track angle and ground track angle, μ i , α i , β i is the roll angle, attack angle and sideslip angle. i ,q i , r i represents the angular velocity of the i-th UAV, is the aerodynamic moment of the i-th UAV; Among them, δ ia , δ ie , δ ir are the aileron, elevator and rudder deflections of the i-th UAV respectively; T i , D i , L i , Y i Indicates thrust, drag, lift and lateral force; Q i ,s i , b i , c i Express dynamic pressure, wing area, wingspan, and mean aerodynamic chord; in, c i2 =(I xi -I yi +I zi )I xzi / ∑ i , c i3 =I zi / Σ i , c i4 =I xzi / Σ i , c i5 =(I zi -I xi ) / I yi , c i6 =I xzi / I yi , c i7 =1 / I yi , c i9 =I xi / ∑ i , I xi , I yi , I zi represents the moment of inertia, I xzi is the moment of inertia, C iD0 , C iDα , C iDα2 , C iY0 , C iYβ , C iL0 , C iLα , C il0 , C ilβ , C ilp , C ilr , C im0 , C imα , C imq , C in0 , C inβ , C inp , and C inr represents the relevant aerodynamic parameters; S32 then converts the dynamic model of the follower UAV i to obtain: Among them, f 1,i , f 2,i , f 3,i , f 4,i , g 1,i , g 2,i , g 3,i , g 4,i represents the corresponding control vector and control matrix, which are derived from the dynamic model of UAV i; X 1,i =[x i ,y i ,h i ] T , X 2,i =[V i ,x i ,γ i ] T , i =[T i ,α i sinμ i ,α i cosμ i ] T , Y 1,i =[μ i ,α i ,β i ] T , Y 2,i =[p i ,q i ,r i ] T ,u i =[δ ia ,δ ie ,δ ir ] T ; At the same time, there are: S33 tracking error e i,1 Taking the second derivative yields: in, By taking the derivative of the dynamic model we get: Design a sliding surface of the following form: Sliding surface S i,1 The derivative is: S34 According to S i,1 Expression and control deviation for UAV cooperative control defined in this application Then design the following virtual control signal: S35 according to the virtual control signal The resilient fault-tolerant collaborative controller of follower UAV i considering link DoS network attack and actuator failure is obtained as:

5. The method for controlling the coordinated resilience of clustered drones against DoS attacks according to claim 4 is characterized in that: According to the virtual control signal Design the intermediate virtual control signal of the follower drone as:

6. The method for controlling the coordinated resilience of clustered drones against DoS attacks according to claim 4 is characterized in that: The resilient fault-tolerant collaborative controller also includes an actuator fault parameter and Designed to estimate and The adaptive law of: The defined variables satisfy the following relationship:

Images