A method for inducing and driving away drone swarms based on deceptive information injection

Abstract

This invention discloses a method for inducing and repelling drone swarms based on deceptive information injection, comprising the following steps: Step 1. Detecting and identifying the drone swarm formation structure and system matrix; Step 2. Designing deceptive information and injecting it into the control loop of the leader drone; Step 3. Determining the theoretical motion trajectory of the leader after the deceptive information injection; Step 4. Determining the theoretical motion trajectory of the followers after the deceptive information injection based on the distributed interaction characteristics of the drone swarm; Step 5. Detecting the actual motion trajectory of the drone swarm after the deceptive information injection; Step 6. Comparing the theoretical and actual motion trajectories to determine whether the drone swarm has been successfully repelled. For drone swarms with a leader-follower structure, this invention injects deceptive information into the leader of the drone swarm and utilizes distributed information interaction between drones to induce and repel the entire drone swarm. This invention does not aim to damage drones, and has the advantages of no collateral damage and a controllable countermeasure effect.

Description

Technical Field

[0001] The present invention relates to the technical field of drone swarm expulsion countermeasures, and in particular to a drone swarm induction and expulsion method based on deception information injection. Background Art

[0002] With the further innovation of artificial intelligence technology, increasingly intelligent and systematic drone swarms have been widely used in social life and military defense fields. Unstable and crashed drone swarms will pose a security threat to personal safety and public safety. Drone swarms controlled by attackers will disrupt aviation order, affect normal aviation flight missions, and cause mission failures. Existing drone swarm countermeasures mainly include hard countermeasures such as destruction and sabotage, and soft countermeasures such as communication link interference. Hard countermeasures such as destruction and sabotage destroy the drone hardware, causing the drone swarm to lose its target or lose its flight capability during flight. Soft countermeasures such as communication link interference mainly interfere with the link signal transmission between drones, affecting the collaborative relationship between drones in the drone swarm, causing its mission to fail. However, during use, the above-mentioned measures still have defects such as low countermeasure cost-effectiveness, poor countermeasure effect, and serious collateral damage.

[0003] The method for inducing and driving away drone swarms by injecting deceptive information mainly focuses on the distributed interaction mechanism of drone swarms. It designs deceptive information and injects it into the control loop of the drone swarm leader, and uses the leader's attraction mechanism to achieve the driving away countermeasure of the entire drone swarm. This method does not aim to destroy drones when used, and the countermeasure effect is controllable, without collateral damage, and there is a possibility of capturing the enemy drone swarm. However, based on existing research results, there is no research on the related methods of driving away drone swarms based on deceptive information injection.

[0004] Based on this, there is an urgent need to design a method for inducing and driving away drone swarms based on deceptive information injection to achieve countermeasures against drone swarms, thereby solving the problems existing in the above-mentioned existing technologies. Summary of the Invention

[0005] In response to the above-mentioned problems, the present invention aims to provide a method for inducing and driving away a swarm of drones based on the injection of deceptive information. This method designs deceptive information in combination with the drone swarm system model, and injects deceptive information into the leader control loop to replace the leader state. The leader's attraction to the followers is utilized to make the leader and followers deviate from the original expected motion trajectory at the same time, thereby achieving the inducing and driving away of the entire drone swarm, and has the characteristics of controllable countermeasure effect and no collateral damage.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] A method for inducing and driving away a group of drones based on deceptive information injection, comprising the steps of:

[0008] Step 1. During the flight of the drone, detect the movement trajectory and distribution characteristics of the drone group, identify the vector composition of the drone formation, and calculate the feasible conditions of the formation. Calculate the system matrix parameter A of the drone swarm;

[0009] Step 2. Design the deception information model matrix H based on the drone swarm system matrix A to obtain the deception information and injecting deceptive information into the control loop of the drone swarm leader;

[0010] Step 3. Obtain the leader’s control protocol and theoretical motion state x after deceptive information injection. N (t);

[0011] Step 4. The drone swarm followers receive the deception information forwarded by the leader and track the deception information to obtain the control protocol and theoretical motion state x of follower i after the deception information is injected. i (t)(i=1,2,…,N-1);

[0012] Step 5. Obtain the actual movement status of the leader and followers of the drone group after the deception information is injected through detection means such as radar and visible light and

[0013] Step 6. Determine whether the actual motion state of the leader and followers after the injection of deceptive information is close to the theoretical motion state. If so, induce the drone swarm to leave. If not, return to Step 2.

[0014] Preferably, the calculation process of the system matrix parameter A of the drone group described in Step 1 includes:

[0015] Step 101. During the flight of the drone group, collect information such as the position, speed, spatial distribution and movement trend of each drone in the drone group, and calculate the relative state difference between the drones. Constructing drone swarm formation vectors

[0016]

[0017] Among them, h ip (t) is the desired position that UAV i needs to maintain in the group, h iv (t) is the desired speed to be maintained;

[0018] Step 102. In the UAV group formation vector, according to the feasible conditions of the UAV group time-varying formation The calculated UAV swarm system matrix A is:

[0019]

[0020] Preferably, the deceptive information is obtained in step 2 The process includes

[0021] Step 201. Deceptive Information The model matrix H in the equation must have the same modality as the drone swarm system matrix A, that is, the deception information must meet the following feasible conditions for expulsion and countermeasures:

[0022]

[0023] That is, the deception information injected into the drone swarm must be based on the status signal of the drone swarm system matrix, otherwise the drone swarm will not be able to achieve formation tracking and expel countermeasures;

[0024] Step 202. Deception information After being injected into the leader control loop, the deceptive information Will directly replace the state information x in the leader control loop N (t), and send deceptive information to the followers in the drone group

[0025] Step 203. Use directional transmitting antenna to transmit deceptive information Injected into the control loop of the drone swarm leader.

[0026] Preferably, the control protocol and theoretical motion state x in step 3 are N The calculation process of (t) includes

[0027] Step 301. Obtain the control protocol of the leader after the deceptive information is injected;

[0028] Step 302. Calculate the theoretical motion state x according to the control protocol of the leader after the deception information is injected N (t).

[0029] Preferably, the calculation process of the control protocol described in step 301 includes:

[0030] (1) Assume that in the distributed UAV swarm dynamics model, there is a leader and N-1 followers. The dynamics model of each UAV is a double integrator system as follows:

[0031]

[0032] Where, l = 1, 2, ..., N, p l (t), v l (t) and u l(t) represents the position, velocity and control input of the lth UAV respectively;

[0033] (2) According to the dual integrator system, the dynamic model of the UAV swarm is obtained as follows:

[0034]

[0035] in, and Denote the states of the leader and follower i respectively, and n denotes the dimension of the drone state vector. For the convenience of description, let n = 1. The conclusion obtained under n = 1 can be extended to multidimensional space through Kronecker product operation. and denote the control inputs of the leader and follower i respectively;

[0036] (3) Construct a directed topological graph G to describe the inter-machine communication interaction network. Assume that each drone corresponds to a node in the topological graph G. The communication interaction relationship between drones is represented by the edge in the topological graph, and the communication interaction intensity is represented by the edge weight w. ij W = [w ij ] N×N and D = diag{d1,d2,…,d N} represent the adjacency matrix and in-degree matrix of the topological graph respectively, where N i represents the set of neighboring drones communicating with the i-th drone; the Laplacian matrix of the topological graph G is L = DW:

[0037]

[0038] in, Represents the communication interaction relationship between followers, Represents the communication interaction between leaders and followers;

[0039] (4) The formation tracking control protocol based on the time-varying state of the UAV swarm is constructed as follows:

[0040]

[0041] Where i = 1, 2, ..., N-1, and The control gain matrix that enables the UAV swarm dynamics model (2) to achieve stable formation tracking; is the expected formation vector that UAV i needs to achieve, is the compensation vector for time-varying formation tracking, c(t)∈R n The motion trajectory that needs to be tracked by the drone swarm;

[0042] (5) This method is based on the stable formation control of UAV swarm. First, the control gain matrix K1 is selected by making A+BK1 a Hurwitz matrix; then the UAV swarm system performance gain matrix Q is given. T =Q>0 and R T =R>0, and the control gain matrix parameter P is obtained by solving the following algebraic Riccati equation:

[0043] A T P+PA-PBR -1 B T P+Q=0

[0044] (6) Finally, K2 = -R -1 B T P obtains the control gain matrix K2. Under the action of the formation control gain matrices K1 and K2, the UAV swarm system realizes time-varying formation tracking.

[0045] Preferably, the calculation of the theoretical motion state x in step 302 is N (t) The process includes

[0046] (1) According to the control protocol in the drone swarm leader’s own control loop:

[0047] u N (t) = K1(x N (t)-c(t))

[0048] When deceptive information x N (t) After injecting and replacing the leader state information, the leader's control protocol is:

[0049]

[0050] (2) Assuming that the time of injection of deceptive information is t1, when t>t1, the leader is affected by the control protocol (3) after the injection of deceptive information, and the theoretical motion state of the leader is:

[0051]

[0052] From formula (4), we can see that the leader deviates from the original motion state and tends to the deceptive state.

[0053] Preferably, the control protocol and theoretical motion state x of follower i after deceptive information injection in step 4 are obtained. i The process of (t)(i=1,2,…,N-1) includes

[0054] Step 401. According to the drone swarm time-varying state formation tracking control protocol, the original control protocol in the drone swarm follower control loop is as follows:

[0055]

[0056] When the drone swarm follower receives the deceptive information x forwarded by the leader N After (t), the actual control protocol of the follower is rewritten as:

[0057]

[0058] Step 402. Based on the UAV swarm follower dynamics model and the control protocol (6) after injecting deceptive information, the theoretical motion state of the follower is described by the following formula:

[0059]

[0060] Followers can still form the expected time-varying formation h i (t)(i=1,2,…,N-1), but the formation tracking trajectory changes, and the follower formation no longer tracks c(t), but tracks the injected deception information x N (t).

[0061] Preferably, the actual motion state of the drone group leader and followers after the deception information injection is obtained in step 5 and The process includes

[0062] Step 501. After the deceptive information is injected, firstly according to the sliding time window expression Set the duration of capturing the motion state of the drone group, i.e. Δt;

[0063] Step 502. In different sliding time windows In the process, the real motion status of each drone in the drone group is tracked and intercepted.

[0064] Preferably, the process of determining whether the actual motion state of the leader and the follower after the deceptive information is injected is close to the theoretical motion state in step 6 includes:

[0065] Step 601. Based on the theoretical motion state x of each drone in the drone group i (t)(i=1,2,…,N) and actual motion state Calculating sliding time windows Inside x i (t) and The error value between:

[0066]

[0067] Step 602. Set the error threshold between the ideal motion state and the actual motion state of the drone group based on the drone type, scene situation and total mission duration. like Then it is determined that the drone group has failed to be driven away, and return to Step 2. If After the deception information is injected, each drone in the drone swarm tracks the motion trajectory corresponding to the deception information, successfully achieving the induction and expulsion of the drone swarm.

[0068] The beneficial effects of the present invention are as follows: the present invention discloses a method for inducing and driving away a swarm of drones based on deceptive information injection. Compared with the prior art, the present invention has the following improvements:

[0069] (1) This paper designs a method for inducing and driving away drone swarms based on deceptive information injection. This method utilizes the distributed interaction mechanism of drone swarms to inject deceptive information into the control loop of the leader of the drone swarm. Combined with the leader's attraction to followers, this method can achieve a countermeasure to drive away the entire drone swarm.

[0070] (2) At the same time, this method aims to drive away the enemy drone swarm with its trajectory. The counter-effect is controllable, there is no collateral damage, and there is the possibility of capturing the enemy drone swarm. It can achieve the induction and expulsion of the entire drone swarm, and has the advantages of controllable counter-effect and no collateral damage. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a flow chart of the method for inducing and driving away a swarm of drones based on deceptive information injection according to the present invention.

[0072] Figure 2 This is a topological diagram of the communication interaction of the drone group in Example 1 of the present invention.

[0073] Figure 3 This is a diagram of the movement trajectory of the drone group before the injection of deceptive information in Example 1 of the present invention.

[0074] Figure 4 This is a diagram of the movement trajectory of the drone group after deceptive information is injected according to Example 1 of the present invention. DETAILED DESCRIPTION

[0075] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0076] Example 1: Refer to the attached Figure 1-4 The method for inducing and driving away a drone group based on deceptive information injection shown in the figure includes the following steps:

[0077] Step 1. During the flight of the drone, detect the movement trajectory and distribution characteristics of the drone group, identify the vector composition of the drone formation, and calculate the feasible conditions of the formation. Calculate the system matrix parameter A of the drone swarm; the specific process includes

[0078] Step 101. During the flight of the drone group, collect information such as the position, speed, spatial distribution and movement trend of each drone in the drone group, and calculate the relative state difference between the drones. Constructing drone swarm formation vectors

[0079]

[0080] Among them, h ip (t) is the desired position that UAV i needs to maintain in the group, h iv (t) is the desired speed to be maintained;

[0081] Step 102. In the UAV group formation vector, according to the feasible conditions of the UAV group time-varying formation The calculated UAV swarm system matrix A is:

[0082]

[0083] Step 2. Design the deception information model matrix H based on the drone swarm system matrix A to obtain the deception information And inject deceptive information into the control loop of the drone swarm leader; the specific process includes

[0084] Step 201. Deceptive Information The model matrix H in the equation must have the same modality as the drone swarm system matrix A, that is, the deception information must meet the following feasible conditions for expulsion and countermeasures:

[0085]

[0086] That is, the deception information injected into the drone swarm must be based on the status signal of the drone swarm system matrix, otherwise the drone swarm will not be able to achieve formation tracking and expel countermeasures;

[0087] Step 202. Deception information After being injected into the leader control loop, the deceptive information Will directly replace the state information x in the leader control loop N (t), and send deceptive information to the followers in the drone group

[0088] Step 203. Use directional transmitting antenna to transmit deceptive information Injection into the control loop of the drone swarm leader;

[0089] Step 3. Obtain the leader’s control protocol and theoretical motion state x after deceptive information injection. N (t); The specific process includes

[0090] Step 301. Obtain the control protocol of the leader after the deceptive information is injected;

[0091] (1) For a swarm of small drones, navigation deception information is designed and injected to make the swarm maintain its formation while deviating from the original preset flight trajectory, thus achieving safe and controllable induced drive-off countermeasures. This method is based on a distributed swarm dynamics model consisting of a leader and N-1 followers. The dynamics model of each drone can be described by the following double integrator system:

[0092]

[0093] Where l = 1, 2, ..., N, p l (t), v l (t) and u l (t) represents the position, velocity and control input of the lth UAV respectively;

[0094] (2) According to the dual integrator system, the dynamic model of the UAV swarm is obtained as follows:

[0095]

[0096] in, and Denote the states of the leader and follower i respectively, and n denotes the dimension of the drone state vector. For the convenience of description, let n = 1. The conclusion obtained under n = 1 can be extended to multidimensional space through Kronecker product operation. and denote the control inputs of the leader and follower i respectively;

[0097] (3) Construct a directed topological graph G to describe the inter-machine communication interaction network. Assume that each drone corresponds to a node in the topological graph G. The communication interaction relationship between drones is represented by the edge in the topological graph, and the communication interaction intensity is represented by the edge weight w. ij W = [w ij ] N×N and D = diag{d1,d2,…,d N} represent the adjacency matrix and in-degree matrix of the topological graph respectively, where N i represents the set of neighboring drones communicating with the i-th drone; the Laplacian matrix of the topological graph G is L = DW, specifically:

[0098]

[0099] in, Represents the communication interaction relationship between followers, Represents the communication interaction between leaders and followers;

[0100] (4) The formation tracking control protocol based on the time-varying state of the UAV swarm is constructed as follows:

[0101]

[0102] Where i = 1, 2, ..., N-1, and The control gain matrix that enables the UAV swarm dynamics model (2) to achieve stable formation tracking; is the expected formation vector that UAV i needs to achieve, is the compensation vector for time-varying formation tracking, which is determined by the system matrix and the formation vector. n The trajectory that the UAV swarm needs to track is determined by the swarm position, mission situation, and target position;

[0103] (5) This method is based on the stable formation control of UAV swarm. First, the control gain matrix K1 is selected by making A+BK1 a Hurwitz matrix; then the UAV swarm system performance gain matrix Q is given. T =Q>0 and R T =R>0, and the control gain matrix parameter P is obtained by solving the following algebraic Riccati equation:

[0104] A T P+PA-PBR -1 B T P+Q=0

[0105] (6) Finally, K2 = -R -1 B T P obtains the control gain matrix K2. Under the action of the formation control gain matrices K1 and K2, the UAV swarm system realizes time-varying formation tracking.

[0106] Step 302. Calculate the theoretical motion state x according to the control protocol of the leader after the deception information is injected N (t);

[0107] (1) According to the control protocol in the drone swarm leader’s own control loop:

[0108] u N (t) = K1(x N (t)-c(t))

[0109] It can be seen that when the deceptive information x N (t) After injecting and replacing the leader state information, the leader's control protocol can be rewritten as:

[0110]

[0111] (2) Assuming that the time of injection of deceptive information is t1, when t>t1, the leader is affected by the control protocol (3) after the injection of deceptive information, and the theoretical motion state of the leader is:

[0112]

[0113] From formula (4), we can see that the leader deviates from the original motion state and tends to the deceptive state;

[0114] Step 4. The drone swarm followers receive the deception information forwarded by the leader and track the deception information to obtain the control protocol and theoretical motion state x of follower i after the deception information is injected. i (t)(i=1,2,…,N-1); the specific process includes

[0115] Step 401. According to the drone swarm time-varying state formation tracking control protocol, the original control protocol in the drone swarm follower control loop is as follows:

[0116]

[0117] When the drone swarm follower receives the deceptive information x forwarded by the leader N After (t), the actual control protocol of the follower can be rewritten as:

[0118]

[0119] Step 402. Based on the UAV swarm follower dynamics model and the control protocol (6) after injecting deceptive information, the theoretical motion state of the follower can be described by the following formula:

[0120]

[0121] It can be seen that the followers can still form the expected time-varying formation h i (t)(i=1,2,…,N-1), but the formation tracking trajectory changes, and the follower formation no longer tracks c(t), but tracks the injected deception information x N (t);

[0122] Step 5. Obtain the actual movement status of the leader and followers of the drone group after the deception information is injected through detection means such as radar and visible light and The specific process includes

[0123] Step 501. After the deceptive information is injected, firstly according to the sliding time window expression Set the duration of capturing the motion state of the drone group, i.e. Δt;

[0124] Step 502. In different sliding time windows In the process of tracking and intercepting the real motion status of each drone in the drone group, including motion trajectory, speed and other information;

[0125] Step 6. Determine whether the actual motion state of the leader and followers after the deception information injection is close to the theoretical motion state. If so, induce the drone group to leave. If not, return to Step 2. The specific process includes

[0126] Step 601. Based on the theoretical motion state x of each drone in the drone group i (t)(i=1,2,…,N) and actual motion state Calculating sliding time windows Inside x i (t) and The error value between:

[0127]

[0128] Step 602. Set the error threshold between the ideal motion state and the actual motion state of the drone group based on the drone type, scene situation and total mission duration. like Then it is determined that the drone group has failed to be driven away, and return to Step 2. If After the deception information was injected, each drone in the drone swarm tracked the motion trajectory corresponding to the deception information, successfully achieving the induced expulsion of the drone swarm.

[0129] Example 2: Different from the above-mentioned Example 1, in order to verify the effectiveness of the technical solution of the above-mentioned Example 1, a simulation experiment of this example is designed to verify the above-mentioned method:

[0130] (1) Consider a UAV swarm system consisting of a leader and five followers, where the dynamics model of each UAV satisfies

[0131]

[0132] The original expected trajectory of the drone group is c(0)=[0,3,0,0] T ; The topology of the drone swarm is as follows Figure 2 As shown, the adjacency weight of the topology graph is selected as 0-1;

[0133] (2) The initial states of the followers and leaders in the drone swarm are:

[0134] x f1 =[5,0,0,0] T ,x f2 =[7,0,5,0] T

[0135] x f3 =[5,0,-5,0] T ,x f4 =[3,0,10,0] T

[0136] x f5 =[5,0,-10,0] T ,x leader =[2,0,5,0] T

[0137] According to the formation feasibility conditions, the formation vectors of the five followers are set to

[0138]

[0139] Where i = 1, 2, ..., 5;

[0140] (3) Selecting the parameter Q = R = I2, the UAV group formation control gain matrix can be solved as follows:

[0141]

[0142]

[0143] (4) Set the simulation time to 30s, and the drone group moves in a uniform linear motion in the positive direction of the X axis. The simulation results are as follows: Figure 3 As shown in the figure; the circle is the mission target that the drone group needs to track, the diamond represents the leader, and the square represents the follower; Figure 3 It can be seen that the follower drones fly in a circle with the leader as the center, and the entire drone swarm flies along the reference trajectory under the attraction of the mission target;

[0144] On the basis of stable formation tracking of the drone swarm, deception information is injected into the leader. The deception information generation system model matrix H is:

[0145]

[0146] When the drone group flies for 20 seconds, the initial state of the deception information is set to Based on the motion state of the drone group, deception information in the positive direction of the Y axis is applied. The flight trajectory of the drone group after injecting the deception information is as follows: Figure 4As shown in the figure, the leader moves away from the original mission target under the influence of deceptive information, and the followers maintain the formation and move in the direction of the deceptive information, achieving the overall induced expulsion countermeasure of the drone group.

[0147] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for inducing and driving away drone swarms based on deceptive information injection, characterized by: Includes steps Step 1. During the flight of the drone, detect the movement trajectory and distribution characteristics of the drone group, identify the vector composition of the drone formation, and calculate the feasible conditions of the formation. , , calculate the system matrix parameters of the drone swarm A ; Step 2. Based on the drone swarm system matrix A Design Deception Information Model Matrix H , get deceptive information and injecting deceptive information into the control loop of the drone swarm leader; Step 3. Obtain the leader's control protocol and theoretical motion state after deceptive information injection ; Step 4. The drone swarm followers receive the deceptive information forwarded by the leader and track the deceptive information to obtain the followers after the deceptive information is injected. i Control protocols and theoretical motion states , ; Step 5. Obtain the actual movement status of the drone group leader and followers after the deception information is injected through radar and visible light detection methods and , ; Step 6. Determine whether the actual motion state of the leader and followers after the deceptive information injection is close to the theoretical motion state. If so, induce the drone swarm to move away. If not, return to Step 2. The process of determining whether the actual motion state of the leader and follower after the injection of deceptive information is close to the theoretical motion state in step 6 includes: Step 601. Based on the theoretical motion state of each drone in the drone swarm and actual motion status , , calculate the sliding time window Inside and The error value between: (11) Step 602. Set the error threshold between the ideal motion state and the actual motion state of the drone group based on the drone type, scene situation and total mission duration. ,like , then it is determined that the drone group has failed to be driven away, and return to Step 2. If After the deception information is injected, each drone in the drone swarm tracks the motion trajectory corresponding to the deception information, and the drone swarm is successfully induced and driven away.

2. The method for inducing and driving away a swarm of drones based on deceptive information injection according to claim 1, characterized in that: The system matrix parameters of the drone swarm described in Step 1 A The calculation process includes Step 101. During the flight of the drone group, collect the position, speed, spatial distribution and movement trend information of each drone in the drone group, and calculate the relative state difference between the drones. , construct the drone group formation vector ； in, For drones i The desired position to be maintained in the fleet, is the desired speed to be maintained; Step 102. In the UAV group formation vector, according to the feasible conditions of the UAV group time-varying formation , the UAV swarm system matrix is calculated A for: (4)。 3. The method for inducing and driving away a swarm of drones based on deceptive information injection according to claim 1, characterized in that: Obtaining deceptive information as described in Step 2 The process includes Step 201. Deceptive Information The model matrix in H Need to be integrated with the drone swarm system matrix A The deceptive information must have the same modality, that is, it must meet the following conditions for feasibility of expulsion countermeasures: (5) That is, the deception information injected into the drone swarm must be based on the status signal of the drone swarm system matrix, otherwise the drone swarm will not be able to achieve formation tracking and expel countermeasures; Step 202. Deception information After being injected into the leader control loop, the deceptive information Will directly replace the state information in the leader control loop , and send deceptive messages to followers in the drone swarm ; Step 203. Use directional transmitting antenna to transmit deceptive information Injected into the control loop of the drone swarm leader.

4. The method for inducing and driving away a swarm of drones based on deceptive information injection according to claim 1, characterized in that: Control protocol and theoretical motion state described in Step 3 The calculation process includes Step 301. Obtain the control protocol of the leader after the deceptive information is injected; Step 302. Calculate the theoretical motion state based on the leader's control protocol after the deceptive information is injected .

5. The method for inducing and driving away a swarm of drones based on deceptive information injection according to claim 4, characterized in that: The calculation process of the control protocol described in step 301 includes: (1) In the distributed UAV swarm dynamics model, there is a leader and The dynamic model of each UAV is a double integrator system as follows: (1) in, , 、 and Respectively represent l The position, velocity, and control inputs of the drone; (2) According to the dual integrator system, the dynamic model of the UAV swarm is obtained as follows: (2) in, and , , representing the leader and follower respectively i status, n Represents the dimension of the drone state vector; for the convenience of description, let , The conclusions obtained in this case can be extended to multi-dimensional space through Kronecker product operation; and Represents leaders and followers respectively i Control input; (3) Constructing a directed topological graph G To describe the inter-machine communication interaction network, let each drone correspond to the topology graph G The communication interaction relationship between drones is represented by the edge in the topology graph, and the communication interaction intensity is represented by the edge weight. express; and Represent the adjacency matrix and in-degree matrix of the topological graph respectively, where , Indicates the i The set of neighboring drones that the drone communicates with; topology map G The Laplacian matrix of for: in, Represents the communication interaction relationship between followers, Represents the communication interaction between leaders and followers; (4) The formation tracking control protocol based on the time-varying state of the UAV swarm is constructed as follows: （3） in, , and To achieve stable formation tracking control gain matrix for the UAV swarm dynamics model (2); , , for drones i The desired formation vector to be achieved, is the compensation vector for time-varying formation tracking, The motion trajectory that needs to be tracked by the drone swarm; (5) Based on the stable formation control of the UAV swarm, first make is the Hurwitz matrix, and the control gain matrix is selected ; Then given the UAV swarm system performance gain matrix and , by solving the following algebraic Riccati equation, we obtain the control gain matrix parameters P : (6) Finally, Get the control gain matrix , in the formation control gain matrix and Under this effect, the drone swarm system realizes time-varying formation tracking.

6. The method for inducing and driving away a swarm of drones based on deceptive information injection according to claim 5, characterized in that: The calculation of the theoretical motion state described in step 302 The process includes (1) According to the control protocol in the drone swarm leader’s own control loop: When deceptive information After injecting and replacing the leader state information, the leader's control protocol is: (6) (2) Assume that the deceptive information injection time is , then when When the leader is affected by the control protocol (3) after the deceptive information is injected, the theoretical motion state of the leader is: (7) From formula (4), we can see that the leader deviates from the original motion state and tends to the deceptive state.

7. The method for inducing and driving away a swarm of drones based on deceptive information injection according to claim 1, characterized in that: Step 4: Obtain the follower after the deceptive information is injected i Control protocols and theoretical motion states The process includes Step 401. According to the drone swarm time-varying state formation tracking control protocol, the original control protocol in the drone swarm follower control loop is as follows: (8) When the drone swarm followers receive deceptive information forwarded by the leader Afterwards, the follower's actual control protocol is rewritten as: (9) Step 402. Based on the UAV swarm follower dynamics model and the control protocol (6) after injecting deceptive information, the theoretical motion state of the follower is described by the following formula: (10) Followers can still form the desired time-varying formation , , but the formation tracking trajectory changes and the follower formation no longer tracks , but instead tracks the injected deceptive information .

8. The method for inducing and driving away a swarm of drones based on deceptive information injection according to claim 1, characterized in that: The actual motion status of the drone group leader and followers after the deception information injection is obtained as described in Step 5 and The process includes Step 501. After the deceptive information is injected, firstly according to the sliding time window expression , set the interception time of the UAV group motion state, that is ; Step 502. In different sliding time windows In the process, the real motion status of each drone in the drone group is tracked and intercepted.

Images