Abnormal individual identification method of drone cluster based on safety set consistency iteration

Abstract

The present invention discloses a method for identifying abnormal individuals in a drone cluster based on consistency iteration of a security set, which belongs to the field of abnormal drone identification and includes the following steps: S1. Initially set the security set for identifying each drone individual in the drone cluster to obtain initial security information and iteration information; S2. Convert the problem of identifying abnormal individuals in the drone cluster into the problem of searching a communication topology graph through the communication topology structure relationship of the drone cluster; S3. Consistency judgment: perform consistency judgment on the content of the security set for identifying each drone in the drone cluster, and output the identification result of the abnormal individual. The present invention adopts the above-mentioned method for identifying abnormal individuals in a drone cluster based on consistency iteration of a security set, which can accurately identify abnormal individuals in a drone cluster, improve the coordination and cooperation capabilities between intelligent bodies, and enhance the adaptability of the system to complex environments, thereby achieving more efficient, robust and reliable system operation.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormal drone identification, and in particular to a method for identifying abnormal individuals in a drone cluster based on security set consistency iteration. Background Art

[0002] Multi-agent systems are complex systems composed of multiple intelligent entities that interact, collaborate, and compete with each other in a shared environment. However, the environments in which multi-agent systems operate are often complex, making them susceptible not only to environmental disturbances but also to external malicious attacks. These attacks are often highly malicious, and if the multi-agent system lacks effective response mechanisms, they can disrupt the consistency of the entire system and even cause irreversible damage.

[0003] Currently, some researchers have proposed detection methods that utilize local or adjacent node information.

[0004] For example, Boem F, Gallo AJ, Ferrari-Trecate G et al. proposed a detection strategy under specific detectable conditions in "A distributed attack detection method for multi-agent systems governed by consensus-based control". However, this strategy cannot detect attacks with certain characteristics.

[0005] On the other hand, Mousavinejad E, Ge X, Han QL, et al. proposed a filter-based detection method in "Detection of cyberattacks on leader-following multi-agent systems." This method uses a filter to generate an ellipsoid state prediction set and an ellipsoid state estimation set. The method then determines whether an attack can be detected by determining whether these two sets intersect. While this detection method only requires local or adjacent node information, it still has some limitations. Summary of the Invention

[0006] To solve the above problems, the present invention provides a method for identifying abnormal individuals in a drone cluster based on security set consistency iteration, so as to accurately identify abnormal individuals in a drone cluster. The method aims to improve the coordination and cooperation capabilities between intelligent agents by accurately identifying abnormal individuals in a drone cluster, and enhance the system's adaptability to complex environments, thereby improving the performance of the multi-agent system in the overall cluster empowerment scenario and achieving more efficient, robust and reliable system operation.

[0007] To achieve the above objectives, the present invention provides a method for identifying abnormal individuals in a drone cluster based on safety set consistency iteration, comprising the following steps:

[0008] S1. Initialization:

[0009] Initially set the security set for each drone in the drone cluster to obtain initial security information and iteration information;

[0010] S2. Update the security collection:

[0011] Through the communication topology relationship of the drone cluster, the problem of identifying abnormal individuals in the drone cluster is transformed into the problem of searching the communication topology graph;

[0012] S3. Consistency judgment:

[0013] The consistency of the security set content identified by each drone in the drone cluster is judged, and the identification results of abnormal individuals are output.

[0014] Preferably, in step S1, each drone in the drone cluster inserts its own individual number into its own security set to obtain initial security information and iteration information.

[0015] Preferably, step S1 specifically includes the following steps:

[0016] S11. Assume that there are a total of drones, of which there are abnormal drones In the drone swarm The drones are numbered as follows: and place the first The secure set of drones identified is denoted as ;

[0017] S12, the first Secure collection of drone identification To perform initial setup:

[0018]

[0019] And record the number of iterations as:

[0020]

[0021] in, The corresponding number of the individual drones in the drone cluster.

[0022] Preferably, in step S2, the neighbor drone safety set information of each drone in the drone cluster is queried, the neighbor drone safety set information is inserted into the safety set of each drone itself, and the entire drone cluster communication topology diagram is traversed to form the final safety information.

[0023] Preferably, step S2 specifically includes the following steps:

[0024] S21. Based on the theory of communication topology graphs, given the following assumptions about the anomaly identification problem in a drone cluster:

[0025] Undirected graph for information communication in drone swarm system network Indicates that is a node set, is an edge set, for For the elements in;

[0026] Adjacency Matrix Defined as:

[0027]

[0028] Assume that the drone cluster communication topology graph is an undirected graph and is connected;

[0029] S22. Convert the problem of identifying abnormal individuals in a drone swarm into the problem of searching the communication topology graph:

[0030] Set the number of drones in the swarm Neighborhood drone safety collection of drones included for:

[0031]

[0032] in, Indicates drone Neighbor drones;

[0033] And set the first The first drone The identified safe set at iteration for:

[0034]

[0035] in, Respectively represent The first drone Element 1 and Element 2 in the security set identified at iteration ;

[0036] The first The first drone The safe set of elements of the identified safe set at iteration for:

[0037]

[0038] in, , , Represent the neighbors of the i-th drone ,Neighbor The security set identified at the n-1th iteration;

[0039] Therefore, the first The first drone The identified safe set at iteration for:

[0040] .

[0041] Preferably, in step S3, the consistency of the safety set information of each individual drone in the drone cluster is determined to end the iterative process and generate an identification result of the abnormal individual.

[0042] Preferably, step S3 specifically includes the following steps:

[0043] S31. Determine the consistency of the security collection information of each individual drone in the drone cluster:

[0044] Set the first The first drone The safe set at the iteration is:

[0045]

[0046] Assume that the first All safe sets at the iteration are:

[0047]

[0048] Determine whether the iteration end condition is met. If so, the iteration ends; otherwise, the iteration continues.

[0049] Step 32: Output abnormal individual identification results:

[0050] The set of numbers of all drones in the drone cluster is :

[0051]

[0052] And remember that the same content of the safety set in the drone cluster is ;

[0053] The identification result of abnormal individuals The result of negating the safety set in the drone cluster:

[0054] .

[0055] Preferably, the iteration end condition described in step S31 is: when the security set elements identified by the drones in the drone cluster maintain consistent security set content during at least two rounds of iterations, the iteration is ended and the consistency judgment flag is recorded as 1.

[0056] The present invention has the following beneficial effects:

[0057] It can accurately identify abnormal individuals in drone clusters, improve the coordination and cooperation capabilities between intelligent agents, and enhance the system's adaptability to complex environments, thereby achieving more efficient, robust and reliable system operation.

[0058] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flow chart of the method for identifying abnormal individuals in a drone cluster based on safety set consistency iteration of the present invention;

[0060] Figure 2 This is a schematic diagram of the communication topology of 0 abnormal individuals described in the simulation experiment 1 of the present invention;

[0061] Figure 3 This is a schematic diagram of the communication topology of an abnormal individual described in Simulation Experiment 2 of the present invention;

[0062] Figure 4 This is a schematic diagram of the communication topology of two abnormal individuals described in simulation experiment 3 of the present invention;

[0063] Figure 5 This is a schematic diagram of the communication topology of three abnormal individuals described in simulation experiment 4 of the present invention. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solutions and advantages disclosed in the embodiments of the present invention clearer, the embodiments of the present invention are further described in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present invention and are not intended to limit the embodiments of the present invention. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, where the same or similar numbers throughout represent the same or similar elements or elements with the same or similar functions.

[0065] It should be noted that the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products or devices.

[0066] Like reference numerals and letters denote like items in the following drawings, and thus, once an item is defined in one drawing, it does not require further definition or explanation in subsequent drawings.

[0067] like Figure 1 As shown in FIG, a method for identifying abnormal individuals in a drone cluster based on safety set consistency iteration includes the following steps:

[0068] S1. Initialization:

[0069] Initially set the security set for each drone in the drone cluster to obtain initial security information and iteration information;

[0070] In step S1, each drone in the drone cluster inserts its own individual number into its own security set to obtain initial security information and iteration information.

[0071] Step S1 specifically includes the following steps:

[0072] S11. Assume that there are a total of drones, of which there are abnormal drones In the drone swarm The drones are numbered as follows: and place the first The secure set of drones identified is denoted as ;

[0073] S12, the first Secure collection of drone identification To perform initial setup:

[0074]

[0075] And record the number of iterations as:

[0076]

[0077] in, is the number of the individual drones in the drone cluster. In the initial iteration, each drone does not exchange information with its neighboring drones based on the communication topology structure, so it will be the first Secure collection of drone identification Set to own number , and set the number of iterations .

[0078] S2. Update the security collection:

[0079] Through the communication topology relationship of the drone cluster, the problem of identifying abnormal individuals in the drone cluster is transformed into the problem of searching the communication topology graph;

[0080] In step S2, the neighbor drone safety set information of each drone in the drone cluster is queried, the neighbor drone safety set information is inserted into the safety set of each drone itself, and the entire drone cluster communication topology diagram is traversed to form the final safety information.

[0081] Step S2 specifically includes the following steps:

[0082] Graph theory is a branch of mathematics that studies the properties and structure of graphs. It is very important for describing the topological structure of communication networks between drones. In a drone cluster, each drone is regarded as a vertex of the graph, and the edges of the graph represent the communication interactions between drones. If two drones can pass messages to each other, it means that there is an edge between them. Conversely, if they cannot pass messages to each other, there is no edge between them. In addition, the direction of communication can be used to construct directed or undirected graphs. If the communication between drone nodes is bidirectional, then the graph is undirected. If the communication is unidirectional, then the graph is directed. In a directed graph, the direction of the edge indicates the direction of message transmission.

[0083] S21. Based on the theory of communication topology graphs, given the following assumptions about the anomaly identification problem in a drone cluster:

[0084] Undirected graph for information communication in drone swarm system network Indicates that is a node set, is an edge set, for For the elements in;

[0085] Adjacency Matrix Defined as:

[0086]

[0087] If there is a path between each pair of drones at different nodes, then the graph is connected, if there is an edge between each pair of different nodes of the drone, then It is a complete graph. Assume that the drone cluster communication topology graph is an undirected graph and is connected;

[0088] S22. Convert the problem of identifying abnormal individuals in a drone swarm into the problem of searching the communication topology graph:

[0089] Set the number of drones in the swarm Neighborhood drone safety collection of drones included for:

[0090]

[0091] in, Indicates drone Neighbor drones;

[0092] And set the first The first drone The identified safe set at iteration for:

[0093]

[0094] in, Respectively represent The first drone Element 1 and Element 2 in the security set identified at iteration ;

[0095] The first The first drone The safe set of elements in the identified safe set at iteration for:

[0096]

[0097] in, , , Represent the neighbors of the i-th drone ,Neighbor The security set identified at the n-1th iteration;

[0098] Therefore, the first The first drone The identified safe set at iteration for:

[0099] .

[0100] It can be seen that the first The first drone The safe set at the iteration is: The first drone The identified security set at the first iteration is the same as the The identified safe set at the iteration corresponds to the safe set union of the elements.

[0101] S3. Consistency judgment:

[0102] The consistency of the security set content identified by each drone in the drone cluster is judged, and the identification results of abnormal individuals are output.

[0103] In step S3, the consistency of the safety set information of each individual drone in the drone cluster is determined to end the iterative process and generate the identification result of the abnormal individual.

[0104] Step S3 specifically includes the following steps:

[0105] S31. Determine the consistency of the security collection information of each individual drone in the drone cluster:

[0106] Set the first The first drone The safe set at the iteration is:

[0107]

[0108] Assume that the first All safe sets at the iteration are:

[0109]

[0110] Determine whether the iteration end condition is met. If so, the iteration ends; otherwise, the iteration continues.

[0111] The iteration end condition described in step S31 is: when the safety set elements identified by the drones in the drone cluster maintain the same safety set content during at least two rounds of iteration, the iteration is ended, and the consistency judgment flag is recorded as 1, and finally the identification of no less than 2 abnormal individuals can be achieved.

[0112] Step 32: Output abnormal individual identification results:

[0113] The set of numbers of all drones in the drone cluster is :

[0114]

[0115] And remember that the same content of the safety set in the drone cluster is ;The same content here refers to the intersection of different drone safety sets;

[0116] The identification result of abnormal individuals The result of negating the safety set in the drone cluster:

[0117] .

[0118] Simulation experiment 1: Assume that the drone cluster contains 0 abnormal individuals and a total of 8 drones. The iterative process is shown in Table 1.

[0119] Table 1

[0120]

[0121] like Figure 2 As shown, there are 5 iterations in total;

[0122] Simulation experiment 2: Assume that the drone cluster contains 1 abnormal individual and a total of 9 drones.

[0123] like Figure 3 As shown, there are 5 iterations in total.

[0124] Simulation experiment 3: Set the drone cluster to contain 2 abnormal individuals and a total of 10 drones. The iterative process is shown in Table 2.

[0125] Table 2

[0126]

[0127] like Figure 4 As shown, there are 6 iterations in total.

[0128] Simulation experiment 4: Set the drone cluster to contain 3 abnormal individuals, a total of 11 drones, and the iterative process is shown in Table 3.

[0129] Table 3

[0130]

[0131] like Figure 5 As shown, there are 6 iterations in total.

[0132] The above simulations illustrate the advantages of the abnormal individual identification method for drone clusters based on safety set consistency iteration described in the present invention, such as low convergence times and fast convergence, and ultimately no less than two abnormal individuals can be identified.

[0133] Therefore, the present invention adopts the above-mentioned drone cluster abnormal individual identification method based on security set consistency iteration, which can accurately identify abnormal individuals in the drone cluster, improve the coordination and cooperation capabilities between intelligent agents, and enhance the system's adaptability to complex environments, thereby achieving more efficient, robust and reliable system operation.

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for identifying abnormal individuals in a drone cluster based on safety set consistency iteration, characterized by: The following steps are involved: S1. Initialization: Initially set the security set for each drone in the drone cluster to obtain initial security information and iteration information; In step S1, each drone in the drone cluster inserts its individual number into its own security set to obtain initial security information and iteration information; Step S1 specifically includes the following steps: S11. Assume that there are a total of drones, of which there are In the drone cluster The drones are numbered as follows: , and the first The safe set of drones identified is denoted as ; S12, the first Safe collection of drone identification To perform initial setup: And record the number of iterations as: in, Corresponding numbers for individual drones in the drone cluster; S2. Update the security collection: Through the communication topology structure of the drone cluster, the problem of identifying abnormal individuals in the drone cluster is transformed into the problem of searching the communication topology graph; In step S2, query the neighbor drone safety set information of each drone in the drone cluster, insert the neighbor drone safety set information into the safety set of each drone itself, and traverse the entire drone cluster communication topology structure diagram to form the final safety information; Step S2 specifically includes the following steps: S21. Based on the theory of communication topology graph, given the assumptions of the anomaly identification problem in drone clusters: Undirected graph for information communication in drone swarm system network Indicates that is a node set, For edge sets, for is the element in; Adjacency Matrix Defined as: Assume that the drone cluster communication topology graph is an undirected graph and is connected; S22. Convert the problem of identifying abnormal individuals in a drone cluster into the problem of searching the communication topology graph: Set the number of drones in the swarm Neighborhood drone safety collection of drones included for: in, Indicates drone Neighborhood drones; And set the first Drone The identified safe set at iteration for: in, Respectively represent Drone Element 1 and 2 in the security set identified at iteration ; The first Drone The safe set of elements in the identified safe set at iteration for: in, , , Represent the neighbors of the i-th drone ,Neighbor The security set identified at the n-1th iteration; Therefore, the first Drone The identified safe set at iteration for: ； S3. Consistency judgment: Make consistency judgment on the security set content of each drone in the drone cluster and output the identification results of abnormal individuals; In step S3, the consistency of the safety set information of each individual drone in the drone cluster is determined to end the iterative process and generate the identification result of the abnormal individual; Step S3 specifically includes the following steps: S31. Determine the consistency of the security collection information of each individual drone in the drone cluster: Set the first Drone The safe set at the iteration is: Suppose the first All safe sets at the iteration are: Determine whether the iteration end condition is met. If so, the iteration ends; otherwise, the iteration continues. The iteration end condition described in step S31 is: when the safety set elements identified by the drones in the drone cluster keep the safety set content consistent in at least two rounds of iterations, the iteration ends and the consistency judgment flag is recorded as 1; Step 32: Output abnormal individual identification results: The serial number set of all drones in the drone cluster is : And remember that the same content of the safety set in the drone cluster is ; The identification result of abnormal individuals The result of negating the safety set in the drone cluster: 。

Images