Fault-tolerant control method and system for cluster unmanned aerial vehicles with malicious faults

Abstract

The invention discloses a cluster unmanned aerial vehicle fault-tolerant control method and system with malicious faults, and relates to the technical field of unmanned aerial vehicle cluster control, and the method comprises the steps: determining the actual value of an acceleration control vector matrix of each unmanned aerial vehicle; determining a malicious unmanned aerial vehicle based on the actual value and the preset value of the acceleration control vector matrix of each unmanned aerial vehicle, and determining a current neighbor unmanned aerial vehicle and a current non-neighbor unmanned aerial vehicle; based on the actual values of the kinematics parameters of the current neighbor unmanned aerial vehicle and each neighbor unmanned aerial vehicle in the corresponding neighbor unmanned aerial vehicle set, determining an optimal value of an acceleration control vector matrix of the current neighbor unmanned aerial vehicle to control the current neighbor unmanned aerial vehicle to fly; and based on the actual values of the kinematics parameters of the current non-neighbor unmanned aerial vehicle and each neighbor unmanned aerial vehicle in the corresponding neighbor unmanned aerial vehicle set, determining the optimal value of the acceleration control vector matrix of the current non-neighbor unmanned aerial vehicle to control the current non-neighbor unmanned aerial vehicle to fly. According to the invention, the problem of fleeing control against malicious unmanned aerial vehicles is solved.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) swarm control technology, and in particular to a fault-tolerant control method and system for swarm UAVs with malicious faults. Background Technology

[0002] With the continuous increase in human economic and social activities, the rapid development of aerospace technology, and its widespread application in the military field, human dependence on aviation is constantly increasing. The mission requirements and complexity of unmanned aerial vehicles (UAVs) are also constantly increasing, with more tasks requiring multiple UAVs to cooperate in a swarm. Therefore, swarm control of UAV swarms has become a hot research topic in the aerospace field. Even meticulously designed and rigorously manufactured UAVs cannot completely avoid various flight malfunctions during flight. The possibility of UAV malfunctions is also increasing. Possible abnormal behaviors in UAVs can be mainly categorized into: physical layer failures, network layer attacks, and abnormal or malicious decisions by the management layer. Common failures at the physical and network layers, such as mechanical failures and communication failures, often lead to a decrease in the accuracy of the entire control system, and in severe cases, even system loss of control.

[0003] In contrast, abnormal or malicious decisions made by management are the most difficult to handle, and can be mainly divided into malicious behavior with clear and modelable decisions, and Byzantine problems with arbitrary and unmodelable decisions. Because this type of behavior differs from traditional failure modes, drones making malicious decisions cannot adjust or compensate for their own malicious impact, rendering general independent fault-tolerant control methods ineffective and leading to the problem of swarm control against malicious drones. Therefore, for drones with clearly defined malicious failures, a fault-tolerant control method for swarm drones with malicious failures is needed to address the problem of swarm control against malicious drones. Summary of the Invention

[0004] The purpose of this application is to provide a fault-tolerant control method and system for swarm drones with malicious faults, in order to solve the problem of resisting the swarm control of malicious drones.

[0005] To achieve the above objectives, this application provides the following solution.

[0006] Firstly, this application provides a fault-tolerant control method for swarmed UAVs with malicious faults, including: Obtain the actual values ​​of the kinematic parameters and the preset values ​​of the acceleration control vector matrix of each UAV in the swarm; the kinematic parameters include: velocity vector and coordinate vector; The set of neighboring drones and the set of non-neighboring drones for each drone are determined based on the configuration of the swarm drones. The actual values ​​of the acceleration control vector matrix of each UAV are determined based on the actual values ​​of the kinematic parameters of each UAV and the actual values ​​of the kinematic parameters of each neighboring UAV in the set of neighboring UAVs. Based on the actual values ​​of the acceleration control vector matrix of each UAV and the preset values ​​of the acceleration control vector matrix of each UAV, malicious UAVs in the UAV cluster are identified; the malicious UAVs are UAVs with malicious malfunctions. Identify any drone in the set of neighboring drones of the malicious drone as the current neighboring drone, and identify any drone in the set of non-neighboring drones of the malicious drone as the current non-neighboring drone; Based on the actual values ​​of the kinematic parameters of the current neighboring drone and the actual values ​​of the kinematic parameters of each neighboring drone in the set of neighboring drones of the current neighboring drone, the optimized value of the acceleration control vector matrix of the current neighboring drone is determined, and the flight of the current neighboring drone is controlled by the optimized value of the acceleration control vector matrix of the current neighboring drone. Based on the actual values ​​of the kinematic parameters of the current non-neighbor UAV and the actual values ​​of the kinematic parameters of each neighbor UAV in the set of neighbor UAVs of the current non-neighbor UAV, the optimized value of the acceleration control vector matrix of the current non-neighbor UAV is determined, and the flight of the current non-neighbor UAV is controlled by the optimized value of the acceleration control vector matrix of the current non-neighbor UAV.

[0007] Secondly, this application provides a fault-tolerant control system for swarm drones with malicious faults, used to implement a fault-tolerant control method for swarm drones with malicious faults. The fault-tolerant control system for swarm drones with malicious faults includes: The data acquisition module is used to acquire the actual values ​​of the kinematic parameters and the preset values ​​of the acceleration control vector matrix of each UAV in the swarm; the kinematic parameters include: velocity vector and coordinate vector; The drone partitioning module is used to determine the set of neighboring drones and the set of non-neighboring drones for each drone based on the configuration of the swarm drones. The actual value determination module is used to determine the actual value of the acceleration control vector matrix of each UAV based on the actual values ​​of the kinematic parameters of each UAV and the actual values ​​of the kinematic parameters of each neighboring UAV in the set of neighboring UAVs of each UAV. The malicious drone identification module is used to identify malicious drones in a drone cluster based on the actual values ​​of the acceleration control vector matrices of each drone and the preset values ​​of the acceleration control vector matrices of each drone; the malicious drones are drones with malicious malfunctions. The drone designation module is used to identify any drone in the set of neighboring drones of a malicious drone as the current neighboring drone, and to identify any drone in the set of non-neighboring drones of a malicious drone as the current non-neighboring drone. The neighbor drone optimization control module is used to determine the optimized value of the acceleration control vector matrix of the current neighbor drone based on the actual values ​​of the kinematic parameters of the current neighbor drone and the actual values ​​of the kinematic parameters of each neighbor drone in the set of neighbor drones of the current neighbor drone, and to control the flight of the current neighbor drone using the optimized value of the acceleration control vector matrix of the current neighbor drone. The non-neighbor drone optimization control module is used to determine the optimized value of the acceleration control vector matrix of the current non-neighbor drone based on the actual values ​​of the kinematic parameters of the current non-neighbor drone and the actual values ​​of the kinematic parameters of each neighbor drone in the set of neighbor drones of the current non-neighbor drone, and to control the flight of the current non-neighbor drone using the optimized value of the acceleration control vector matrix of the current non-neighbor drone.

[0008] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application discloses a fault-tolerant control method and system for swarmed drones with malicious faults. First, based on the actual values ​​of the kinematic parameters of each drone and the actual values ​​of the kinematic parameters of each drone's neighboring drones, the actual values ​​of the acceleration control vector matrix of each drone are determined. Then, based on the actual values ​​of the acceleration control vector matrices of each drone and preset values ​​of the acceleration control vector matrices of each drone, malicious drones in the swarm are identified. Second, any drone in the neighboring drone set of the malicious drone is identified as the current neighboring drone, and any drone in the non-neighboring drone set of the malicious drone is identified as the current non-neighboring drone. Third, ... Based on the actual values ​​of the kinematic parameters of the current neighboring UAV and the actual values ​​of the kinematic parameters of each neighboring UAV in the set of neighboring UAVs of the current neighboring UAV, the optimized value of the acceleration control vector matrix of the current neighboring UAV is determined, and the flight of the current neighboring UAV is controlled using the optimized value of the acceleration control vector matrix of the current neighboring UAV. Finally, based on the actual values ​​of the kinematic parameters of the current non-neighboring UAV and the actual values ​​of the kinematic parameters of each neighboring UAV in the set of neighboring UAVs of the current non-neighboring UAV, the optimized value of the acceleration control vector matrix of the current non-neighboring UAV is determined, and the flight of the current non-neighboring UAV is controlled using the optimized value of the acceleration control vector matrix of the current non-neighboring UAV. This application is based on the configuration of swarm UAVs, which achieves the swarming target of swarm UAVs while retaining malicious UAVs. A layered configuration is constructed with malicious UAVs as the first layer to restrain malicious UAVs. Cooperative control is achieved for the set of neighboring UAVs and the set of non-neighboring UAVs of the malicious UAVs according to the set swarming control method. By controlling the neighboring UAVs of the malicious UAVs to the required geometric shape to restrain the malicious UAVs, fault-tolerant control of swarm UAVs is achieved. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A schematic flowchart of a fault-tolerant control method for a cluster of unmanned aerial vehicles with malicious faults provided in an embodiment of this application; Figure 2 This is a schematic diagram of the three-layer geometric configuration of a swarm of drones. Detailed Implementation

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0012] The purpose of this application is to provide a fault-tolerant control method and system for swarm drones with malicious faults, aiming to solve the problem of resisting the swarm control of malicious drones.

[0013] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0014] In one exemplary embodiment, such as Figure 1 As shown, a fault-tolerant control method for swarmed UAVs with malicious faults is provided, including the following steps.

[0015] Step 1: Obtain the actual values ​​of the kinematic parameters and the preset values ​​of the acceleration control vector matrix of each UAV in the swarm; the kinematic parameters include: velocity vector and coordinate vector.

[0016] Step 2: Determine the set of neighboring drones and the set of non-neighboring drones for each drone based on the configuration of the swarm drones.

[0017] Step 3: Determine the actual value of the acceleration control vector matrix for each UAV based on the actual values ​​of the kinematic parameters of each UAV and the actual values ​​of the kinematic parameters of each neighboring UAV in the set of neighboring UAVs.

[0018] As an optional implementation method, step 3 specifically includes: Using a fixed-velocity swarm control law, the actual values ​​of the acceleration control vector matrix for each UAV are calculated based on the actual values ​​of the kinematic parameters of each UAV and the actual values ​​of the kinematic parameters of each UAV in its neighboring UAV set. The fixed-velocity swarm control law includes: ; ; in, For drones The actual value of the acceleration control vector matrix; For drones A collection of neighboring drones; For drones With drones The actual value of the relative velocity, , For drones The actual value of the velocity vector, For drones The actual value of the velocity vector; To Find the partial derivative. For drones The actual value of the coordinate vector; For swarm drones; For about The potential energy function, For drones With drones The actual value of the relative distance, , For drones The actual value of the coordinate vector; It is the absolute value; For drones With drones The attractive potential energy between them; For drones With drones The repulsive potential energy between them; The sensing radius of the drone; To set a constant, .

[0019] Step 4: Based on the actual values ​​of the acceleration control vector matrix of each UAV and the preset values ​​of the acceleration control vector matrix of each UAV, identify the malicious UAVs in the cluster; malicious UAVs are UAVs with malicious malfunctions.

[0020] As an optional implementation, step 4 specifically includes: Drones whose actual values ​​of the acceleration control vector matrix in a swarm of drones are not equal to the preset values ​​of the acceleration control vector matrix are identified as malicious drones.

[0021] Step 5: Identify any drone in the set of neighboring drones of the malicious drone as the current neighboring drone, and identify any drone in the set of non-neighboring drones of the malicious drone as the current non-neighboring drone.

[0022] Specifically, step 5 resulted in the following: Figure 2 The diagram shows the three-layer geometric configuration of the cluster of UAVs. Figure 2 In this model, a malicious drone (malicious individual) is placed at layer 1, all its neighboring drones (adjacent individuals) are at layer 2, and non-neighboring drones (remaining individuals) are at layer 3. Drones in layer 2 do not utilize information from drones in layer 3. (Definition) The set of drones consisting of all neighboring drones of the malicious drone and its neighboring drones. , For malicious drones The ensemble of drones For malicious drones A collection of neighboring drones, This refers to a collection of non-neighboring drones used by malicious drones. , , and All were malicious drones Neighbor drones , , and All were malicious drones Non-neighbor drones.

[0023] Step 6: Based on the actual values ​​of the kinematic parameters of the current neighboring drone and the actual values ​​of the kinematic parameters of each neighboring drone in the set of neighboring drones of the current neighboring drone, determine the optimized value of the acceleration control vector matrix of the current neighboring drone, and use the optimized value of the acceleration control vector matrix of the current neighboring drone to control the flight of the current neighboring drone.

[0024] As an optional implementation, in step 6, based on the actual values ​​of the kinematic parameters of the current neighboring drone and the actual values ​​of the kinematic parameters of each neighboring drone in the set of neighboring drones of the current neighboring drone, the optimized value of the acceleration control vector matrix of the current neighboring drone is determined, specifically including: Using the neighbor drone control law, based on the actual values ​​of the kinematic parameters of the current neighbor drone and the actual values ​​of the kinematic parameters of each neighbor drone in the set of neighbor drones, the optimized value of the acceleration control vector matrix of the current neighbor drone is calculated; the neighbor drone control law includes: ; ; ; ; ; ; ; ; in, For drones The optimized value of the acceleration control vector matrix; For drones A collection of neighboring drones; The set of drones consisting of all neighboring drones of the malicious drone and its neighboring drone set; and All are preset constants. , ; For drones The actual value of the velocity vector; For drones The actual value of the velocity vector; To Find the partial derivative. For drones The coordinate vector; For about The potential energy function, For drones With drones The actual value of the relative distance, For drones With drones The expected value of the relative distance; For The potential energy function; For malicious drones A collection of neighboring drones; For drones With drones Intermediate calculated values ​​between; This is the intermediate matrix; For drones and Initial values ​​of the fault-tolerant potential energy function between; For drones The initial value of the velocity vector; For malicious drones The initial value of the velocity vector; For transpose; To obtain the minimum value; and All of these are malicious parameters. , ; For malicious drones With drones The attractive potential energy between them; For malicious drones With drones The repulsive potential energy between them.

[0025] Step 7: Based on the actual values ​​of the kinematic parameters of the current non-neighbor UAV and the actual values ​​of the kinematic parameters of each neighbor UAV in the set of neighbor UAVs of the current non-neighbor UAV, determine the optimized value of the acceleration control vector matrix of the current non-neighbor UAV, and use the optimized value of the acceleration control vector matrix of the current non-neighbor UAV to control the flight of the current non-neighbor UAV.

[0026] As an optional implementation, in step 7, based on the actual values ​​of the kinematic parameters of the current non-neighbor UAV and the actual values ​​of the kinematic parameters of each neighbor UAV in the set of neighbor UAVs of the current non-neighbor UAV, the optimized value of the acceleration control vector matrix of the current non-neighbor UAV is determined, specifically including: Using the non-neighbor UAV control law, based on the actual values ​​of the kinematic parameters of the current non-neighbor UAV and the actual values ​​of the kinematic parameters of each neighbor UAV in the set of neighbor UAVs of the current non-neighbor UAV, the optimized value of the acceleration control vector matrix of the current non-neighbor UAV is calculated; the non-neighbor UAV control law includes: ; ; ; in, For drones The optimized value of the acceleration control vector matrix; For drones A collection of neighboring drones; For drones With drones Time-varying gain; For the sgn function; For drones The actual value of the velocity vector; For drones The actual value of the velocity vector; To Find the partial derivative. For drones The actual value of the coordinate vector; For about The potential energy function; for The derivative; For drones With drones The positive constant between; It is the 1-norm of the vector; For drones With drones The attractive potential energy between them; For drones With drones The repulsive potential energy between them; For drones With drones The actual value of the relative distance, , For drones The actual value of the coordinate vector.

[0027] The main idea of ​​the geometry-based controller in this application is to restrain a malicious drone by "pulling" its neighboring drones into a desired geometry. The first objective is to ensure that neighboring drones reach the same common speed; the second is to bring neighboring drones closer to their desired configuration; and the last is to compensate for the impact of malicious drones on their neighboring drones. The last two objectives guarantee that neighboring drones have a greater capacity to achieve their goals than malicious drones.

[0028] In one exemplary embodiment, a fault-tolerant control system for swarm drones with malicious faults is provided, for implementing a fault-tolerant control method for swarm drones with malicious faults. The fault-tolerant control system for swarm drones with malicious faults includes: The data acquisition module is used to acquire the actual values ​​of the kinematic parameters of each UAV in the swarm and the preset values ​​of the acceleration control vector matrix; the kinematic parameters include: velocity vector and coordinate vector; The drone partitioning module is used to determine the set of neighboring drones and the set of non-neighboring drones for each drone based on the configuration of the swarm drones. The actual value determination module is used to determine the actual value of the acceleration control vector matrix of each UAV based on the actual values ​​of the kinematic parameters of each UAV and the actual values ​​of the kinematic parameters of each neighboring UAV in the set of neighboring UAVs of each UAV. The malicious drone identification module is used to identify malicious drones in the drone swarm based on the actual values ​​of the acceleration control vector matrices of each drone and the preset values ​​of the acceleration control vector matrices of each drone; malicious drones are drones with malicious malfunctions. The drone designation module is used to identify any drone in the set of neighboring drones of a malicious drone as the current neighboring drone, and to identify any drone in the set of non-neighboring drones of a malicious drone as the current non-neighboring drone. The neighbor drone optimization control module is used to determine the optimized value of the acceleration control vector matrix of the current neighbor drone based on the actual values ​​of the kinematic parameters of the current neighbor drone and the actual values ​​of the kinematic parameters of each neighbor drone in the set of neighbor drones of the current neighbor drone, and to control the flight of the current neighbor drone using the optimized value of the acceleration control vector matrix of the current neighbor drone. The non-neighbor drone optimization control module is used to determine the optimized value of the acceleration control vector matrix of the current non-neighbor drone based on the actual values ​​of the kinematic parameters of the current non-neighbor drone and the actual values ​​of the kinematic parameters of each neighbor drone in the set of neighbor drones of the current non-neighbor drone, and to control the flight of the current non-neighbor drone using the optimized value of the acceleration control vector matrix of the current non-neighbor drone.

[0029] As an optional implementation, the actual value of the acceleration control vector matrix of each UAV is determined based on the actual values ​​of the kinematic parameters of each UAV and the actual values ​​of the kinematic parameters of each neighboring UAV in the set of neighboring UAVs of each UAV. Specifically, this includes: Using a fixed-velocity swarm control law, the actual values ​​of the acceleration control vector matrix for each UAV are calculated based on the actual values ​​of the kinematic parameters of each UAV and the actual values ​​of the kinematic parameters of each UAV in its neighboring UAV set. The fixed-velocity swarm control law includes: ; ; in, For drones The actual value of the acceleration control vector matrix; For drones A collection of neighboring drones; For drones With drones The actual value of the relative velocity, , For drones The actual value of the velocity vector, For drones The actual value of the velocity vector; To Find the partial derivative. For drones The actual value of the coordinate vector; For swarm drones; For about The potential energy function, For drones With drones The actual value of the relative distance, , For drones The actual value of the coordinate vector; It is the absolute value; For drones With drones The attractive potential energy between them; For drones With drones The repulsive potential energy between them; The sensing radius of the drone; To set a constant, .

[0030] As an optional implementation, based on the actual values ​​of the acceleration control vector matrices of each UAV and the preset values ​​of the acceleration control vector matrices of each UAV, malicious UAVs in the swarm are identified, specifically including: Drones whose actual values ​​of the acceleration control vector matrix in a swarm of drones are not equal to the preset values ​​of the acceleration control vector matrix are identified as malicious drones.

[0031] As an optional implementation, the optimized value of the acceleration control vector matrix of the current neighboring UAV is determined based on the actual values ​​of the kinematic parameters of the current neighboring UAV and the actual values ​​of the kinematic parameters of each neighboring UAV in the set of neighboring UAVs of the current neighboring UAV. Specifically, this includes: Using the neighbor drone control law, based on the actual values ​​of the kinematic parameters of the current neighbor drone and the actual values ​​of the kinematic parameters of each neighbor drone in the set of neighbor drones, the optimized value of the acceleration control vector matrix of the current neighbor drone is calculated; the neighbor drone control law includes: ; ; ; ; ; ; ; ; in, For drones The optimized value of the acceleration control vector matrix; For drones A collection of neighboring drones; The set of drones consisting of all neighboring drones of the malicious drone and its neighboring drone set; and All are preset constants. , ; For drones The actual value of the velocity vector; For drones The actual value of the velocity vector; To Find the partial derivative. For drones The coordinate vector; For about The potential energy function, For drones With drones The actual value of the relative distance, For drones With drones The expected value of the relative distance; For The potential energy function; For malicious drones A collection of neighboring drones; For drones With drones Intermediate calculated values ​​between; This is the intermediate matrix; For drones and Initial values ​​of the fault-tolerant potential energy function between; For drones The initial value of the velocity vector; For malicious drones The initial value of the velocity vector; For transpose; To obtain the minimum value; and All of these are malicious parameters. , ; For malicious drones With drones The attractive potential energy between them; For malicious drones With drones The repulsive potential energy between them.

[0032] As an optional implementation, based on the actual values ​​of the kinematic parameters of the current non-neighbor UAV and the actual values ​​of the kinematic parameters of each neighbor UAV in the set of neighbor UAVs of the current non-neighbor UAV, the optimized value of the acceleration control vector matrix of the current non-neighbor UAV is determined, specifically including: Using the non-neighbor UAV control law, based on the actual values ​​of the kinematic parameters of the current non-neighbor UAV and the actual values ​​of the kinematic parameters of each neighbor UAV in the set of neighbor UAVs of the current non-neighbor UAV, the optimized value of the acceleration control vector matrix of the current non-neighbor UAV is calculated; the non-neighbor UAV control law includes: ; ; ; in, For drones The optimized value of the acceleration control vector matrix; For drones A collection of neighboring drones; For drones With drones Time-varying gain; For the sgn function; For drones The actual value of the velocity vector; For drones The actual value of the velocity vector; To Find the partial derivative. For drones The actual value of the coordinate vector; For about The potential energy function; for The derivative; For drones With drones The positive constant between; It is the 1-norm of the vector; For drones With drones The attractive potential energy between them; For drones With drones The repulsive potential energy between them; For drones With drones The actual value of the relative distance, , For drones The actual value of the coordinate vector.

[0033] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0034] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0035] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A cluster unmanned aerial vehicle fault-tolerant control method with malicious faults, characterized in that, The cluster unmanned vehicle fault-tolerant control method with malicious faults comprises: obtaining actual values of kinematic parameters of each unmanned vehicle in the cluster unmanned vehicle and preset values of acceleration control vector matrices of the unmanned vehicles; the kinematic parameters include a velocity vector and a coordinate vector; determining a neighbor unmanned vehicle set and a non-neighbor unmanned vehicle set of each unmanned vehicle based on a configuration of the cluster unmanned vehicle; determining actual values of the acceleration control vector matrices of the unmanned vehicles based on the actual values of the kinematic parameters of each unmanned vehicle and the actual values of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of each unmanned vehicle; determining a malicious unmanned vehicle in the cluster unmanned vehicle based on the actual values of the acceleration control vector matrices of the unmanned vehicles and the preset values of the acceleration control vector matrices of the unmanned vehicles; the malicious unmanned vehicle is an unmanned vehicle with a malicious fault; determining any unmanned vehicle in the neighbor unmanned vehicle set of the malicious unmanned vehicle as a current neighbor unmanned vehicle and determining any unmanned vehicle in the non-neighbor unmanned vehicle set of the malicious unmanned vehicle as a current non-neighbor unmanned vehicle; determining an optimized value of the acceleration control vector matrix of the current neighbor unmanned vehicle based on the actual values of the kinematic parameters of the current neighbor unmanned vehicle and the actual values of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of the current neighbor unmanned vehicle, and controlling the current neighbor unmanned vehicle to fly by using the optimized value of the acceleration control vector matrix of the current neighbor unmanned vehicle; determining an optimized value of the acceleration control vector matrix of the current non-neighbor unmanned vehicle based on the actual values of the kinematic parameters of the current non-neighbor unmanned vehicle and the actual values of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of the current non-neighbor unmanned vehicle, and controlling the current non-neighbor unmanned vehicle to fly by using the optimized value of the acceleration control vector matrix of the current non-neighbor unmanned vehicle.

2. The cluster UAV fault-tolerant control method with malicious faults according to claim 1, characterized in that, The method comprises the following steps: determining actual values of the acceleration control vector matrices of the unmanned vehicles based on the actual values of the kinematic parameters of each unmanned vehicle and the actual values of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of each unmanned vehicle, specifically comprising: ； ； in, For drones The actual value of the acceleration control vector matrix; For drones A collection of neighboring drones; For drones With drones The actual value of the relative velocity, , For drones The actual value of the velocity vector, For drones The actual value of the velocity vector; To Find the partial derivative. For drones The actual value of the coordinate vector; For swarm drones; For about The potential energy function, For drones With drones The actual value of the relative distance, , For drones The actual value of the coordinate vector; It is the absolute value; For drones With drones The attractive potential energy between them; For drones With drones The repulsive potential energy between them; The sensing radius of the drone; To set a constant, .

3. The cluster UAV fault-tolerant control method with malicious faults according to claim 1, characterized in that, calculating the actual values of the acceleration control vector matrices of the unmanned vehicles according to the actual values of the kinematic parameters of each unmanned vehicle and the actual values of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of each unmanned vehicle by using a fixed-speed cluster control law; the fixed-speed cluster control law comprises: determining a malicious unmanned vehicle in the cluster unmanned vehicle based on the actual values of the acceleration control vector matrices of the unmanned vehicles and the preset values of the acceleration control vector matrices of the unmanned vehicles, specifically comprising:

4. The cluster UAV fault-tolerant control method with malicious faults according to claim 2, characterized in that, determining the unmanned vehicle with the actual values of the acceleration control vector matrices different from the preset values of the acceleration control vector matrices as the malicious unmanned vehicle. determining an optimized value of the acceleration control vector matrix of the current neighbor unmanned vehicle based on the actual values of the kinematic parameters of the current neighbor unmanned vehicle and the actual values of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of the current neighbor unmanned vehicle, specifically comprising: The optimization value of the acceleration control vector matrix of the current neighbor unmanned aerial vehicle is calculated according to the actual value of the kinematic parameter of the current neighbor unmanned aerial vehicle and the actual value of the kinematic parameter of each neighbor unmanned aerial vehicle in the neighbor unmanned aerial vehicle set of the current neighbor unmanned aerial vehicle by using a neighbor unmanned aerial vehicle control law; the neighbor unmanned aerial vehicle control law comprises: ； ； ； ； ； ； ； ； wherein, is an acceleration control vector matrix of the UAV ; is a neighbor UAV set of the UAV ; is a UAV set consisting of the malicious UAV and all neighbors of the malicious UAV in the neighbor UAV set of the malicious UAV; and are both preset constants, , ; is an actual value of a velocity vector of the UAV ; is an actual value of a velocity vector of the UAV ; is a partial derivative of , is a coordinate vector of the UAV ; is a potential energy function about , is an actual value of a relative distance between the UAV and the UAV , is an expected value of a relative distance between the UAV and the UAV ; is a potential energy function about ; is a neighbor UAV set of the malicious UAV ; is an intermediate calculation value between the UAV and the UAV ; is an intermediate matrix; is a fault-tolerant potential energy function initial value between the UAV and ; is an initial value of a velocity vector of the UAV ; is an initial value of a velocity vector of the malicious UAV ; is a transpose; is a minimum value; and are both malicious parameters, , ; is an attractive potential energy between the malicious UAV and the UAV ; is a repulsive potential energy between the malicious UAV and the UAV .

5. The cluster UAV fault-tolerant control method with malicious faults according to claim 4, characterized in that, The optimization value of the acceleration control vector matrix of the current non-neighbor unmanned aerial vehicle is determined based on the actual value of the kinematic parameter of the current non-neighbor unmanned aerial vehicle and the actual value of the kinematic parameter of each neighbor unmanned aerial vehicle in the neighbor unmanned aerial vehicle set of the current non-neighbor unmanned aerial vehicle, and specifically comprises: The optimization value of the acceleration control vector matrix of the current non-neighbor unmanned aerial vehicle is calculated according to the actual value of the kinematic parameter of the current non-neighbor unmanned aerial vehicle and the actual value of the kinematic parameter of each neighbor unmanned aerial vehicle in the neighbor unmanned aerial vehicle set of the current non-neighbor unmanned aerial vehicle by using a non-neighbor unmanned aerial vehicle control law; the non-neighbor unmanned aerial vehicle control law comprises: ； ； ； wherein, is an acceleration control vector matrix for the UAV ; is a neighbor UAV set for the UAV ; is a time-varying gain between the UAV and the UAV ; is a sgn function; is an actual value of a velocity vector for the UAV ; is an actual value of a velocity vector for the UAV ; is a partial derivative of , is an actual value of a coordinate vector for the UAV ; is a potential energy function for ; is a derivative of ; is a constant between the UAV and the UAV ; is a 1-norm of a vector; is an attractive potential energy between the UAV and the UAV ; is a repulsive potential energy between the UAV and the UAV ; is an actual value of a relative distance between the UAV and the UAV , , is an actual value of a coordinate vector for the UAV .

6. A cluster UAV fault-tolerant control system with malicious faults, configured to implement the cluster UAV fault-tolerant control method with malicious faults according to any one of claims 1-5, characterized in that, The fault-tolerant control system of the cluster unmanned aerial vehicle with malicious faults comprises: The data acquisition module is configured to acquire the actual values of the kinematic parameters of the unmanned aerial vehicles in the cluster unmanned aerial vehicle and the preset values of the acceleration control vector matrices; the kinematic parameters comprise a velocity vector and a coordinate vector; The unmanned aerial vehicle division module is configured to determine the neighbor unmanned aerial vehicle set and the non-neighbor unmanned aerial vehicle set of each unmanned aerial vehicle based on the configuration of the cluster unmanned aerial vehicle; The actual value determination module is configured to determine the actual values of the acceleration control vector matrices of the unmanned aerial vehicles based on the actual values of the kinematic parameters of the unmanned aerial vehicles and the actual values of the kinematic parameters of the neighbor unmanned aerial vehicles in the neighbor unmanned aerial vehicle set of each unmanned aerial vehicle, respectively. The malicious unmanned aerial vehicle determination module is configured to determine the malicious unmanned aerial vehicle in the cluster unmanned aerial vehicle based on the actual values of the acceleration control vector matrices of the unmanned aerial vehicles and the preset values of the acceleration control vector matrices of the unmanned aerial vehicles; the malicious unmanned aerial vehicle is an unmanned aerial vehicle with malicious faults. The unmanned aerial vehicle designation module is configured to determine any unmanned aerial vehicle in the neighbor unmanned aerial vehicle set of the malicious unmanned aerial vehicle as the current neighbor unmanned aerial vehicle, and determine any unmanned aerial vehicle in the non-neighbor unmanned aerial vehicle set of the malicious unmanned aerial vehicle as the current non-neighbor unmanned aerial vehicle. The neighbor unmanned aerial vehicle optimization control module is configured to determine the optimization value of the acceleration control vector matrix of the current neighbor unmanned aerial vehicle based on the actual value of the kinematic parameter of the current neighbor unmanned aerial vehicle and the actual value of the kinematic parameter of each neighbor unmanned aerial vehicle in the neighbor unmanned aerial vehicle set of the current neighbor unmanned aerial vehicle, and control the flight of the current neighbor unmanned aerial vehicle by using the optimization value of the acceleration control vector matrix of the current neighbor unmanned aerial vehicle. The non-neighbor unmanned aerial vehicle optimization control module is configured to determine the optimization value of the acceleration control vector matrix of the current non-neighbor unmanned aerial vehicle based on the actual value of the kinematic parameter of the current non-neighbor unmanned aerial vehicle and the actual value of the kinematic parameter of each neighbor unmanned aerial vehicle in the neighbor unmanned aerial vehicle set of the current non-neighbor unmanned aerial vehicle, and control the flight of the current non-neighbor unmanned aerial vehicle by using the optimization value of the acceleration control vector matrix of the current non-neighbor unmanned aerial vehicle.

7. The cluster UAV fault-tolerant control system with malicious faults according to claim 6, characterized in that, The actual values of the acceleration control vector matrices of the unmanned aerial vehicles are determined based on the actual values of the kinematic parameters of the unmanned aerial vehicles and the actual values of the kinematic parameters of the neighbor unmanned aerial vehicles in the neighbor unmanned aerial vehicle set of each unmanned aerial vehicle, respectively, and specifically comprise: The actual value of the acceleration control vector matrix of each unmanned vehicle is calculated according to the actual value of the kinematic parameters of each unmanned vehicle and the actual value of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of each unmanned vehicle by using a fixed speed swarm control law, and the fixed speed swarm control law comprises: ； ； wherein, is an acceleration control vector matrix of the UAV ; is a neighbor UAV set of the UAV ; is a relative velocity of the UAV ; , , is a velocity vector of the UAV ; is a velocity vector of the UAV ; is a partial derivative of , is a coordinate vector of the UAV ; is a swarm UAV; is a potential function with respect to , is a relative distance of the UAV ; , , is a coordinate vector of the UAV ; is an absolute value; is an attractive potential energy between the UAV ; , is a repulsive potential energy between the UAV ; , is a UAV sensing radius; is a set constant, .

8. The cluster of UAVs fault-tolerant control system with malicious faults according to claim 6, wherein, Based on the actual value of the acceleration control vector matrix of each unmanned vehicle and the preset value of the acceleration control vector matrix of each unmanned vehicle, a malicious unmanned vehicle in the swarm unmanned vehicle is determined, and specifically comprising: The unmanned vehicle whose actual value of the acceleration control vector matrix in the swarm unmanned vehicle is not equal to the preset value of the acceleration control vector matrix is determined as a malicious unmanned vehicle.

9. The cluster of UAVs fault-tolerant control system with malicious faults according to claim 7, wherein, Based on the actual value of the kinematic parameters of the current neighbor unmanned vehicle and the actual value of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of the current neighbor unmanned vehicle, the optimization value of the acceleration control vector matrix of the current neighbor unmanned vehicle is determined, and specifically comprising: The optimization value of the acceleration control vector matrix of the current neighbor unmanned vehicle is calculated according to the actual value of the kinematic parameters of the current neighbor unmanned vehicle and the actual value of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of the current neighbor unmanned vehicle by using a neighbor unmanned vehicle control law, and the neighbor unmanned vehicle control law comprises: ； ； ； ； ； ； ； ； wherein, is an acceleration control vector matrix of the UAV ; is a neighbor UAV set of the UAV ; is a UAV set consisting of the malicious UAV and all neighbors of the malicious UAV in the neighbor UAV set of the malicious UAV; and are preset constants, , ; is an actual value of a velocity vector of the UAV ; is an actual value of a velocity vector of the UAV ; is a partial derivative of , is a coordinate vector of the UAV ; is a potential energy function about , is an actual value of a relative distance between the UAV and the UAV , is an expected value of a relative distance between the UAV and the UAV ; is a potential energy function about ; is a neighbor UAV set of the malicious UAV ; is an intermediate calculation value between the UAV and the UAV ; is an intermediate matrix; is a fault-tolerant potential energy function initial value between the UAV and ; is an initial value of a velocity vector of the UAV ; is an initial value of a velocity vector of the malicious UAV ; is a transpose; is a minimum value; and are malicious parameters, , ; is an attractive potential energy between the malicious UAV and the UAV ; is a repulsive potential energy between the malicious UAV and the UAV .

10. The cluster of UAVs fault-tolerant control system with malicious faults according to claim 9, characterized in that, Based on the actual value of the kinematic parameters of the current non-neighbor unmanned vehicle and the actual value of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of the current non-neighbor unmanned vehicle, the optimization value of the acceleration control vector matrix of the current non-neighbor unmanned vehicle is determined, and specifically comprising: The optimization value of the acceleration control vector matrix of the current non-neighbor unmanned vehicle is calculated according to the actual value of the kinematic parameters of the current non-neighbor unmanned vehicle and the actual value of the kinematic parameters of each neighbor unmanned vehicle in the neighbor unmanned vehicle set of the current non-neighbor unmanned vehicle by using a non-neighbor unmanned vehicle control law, and the non-neighbor unmanned vehicle control law comprises: ； ； ； in, For drones The optimized value of the acceleration control vector matrix; For drones A collection of neighboring drones; For drones With drones Time-varying gain; For the sgn function; For drones The actual value of the velocity vector; For drones The actual value of the velocity vector; To Find the partial derivative. For drones The actual value of the coordinate vector; For about The potential energy function; for The derivative; For drones With drones The positive constant between; It is the 1-norm of the vector; For drones With drones The attractive potential energy between them; For drones With drones The repulsive potential energy between them; For drones With drones The actual value of the relative distance, , For drones The actual value of the coordinate vector.

Images