Unmanned aerial vehicle cluster emergence behavior identification method and system

By identifying the emergence behavior of the drone cluster based on the probability distribution difference measurement method based on the discrete divergence, and determining the intervention nodes using the reverse control model, the problem of difficulty in revealing the operation rules mechanism of the drone cluster and identifying fragile nodes is solved, and the problem of the difficulty in revealing the operation rules mechanism of the drone cluster is realized. Effective identification and reverse control of drone clusters.

CN119088078APending Publication Date: 2024-12-06CHINESE PEOPLES LIBERATION ARMY UNIT 32180
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411184808.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing technology is difficult to effectively reveal the operating rules and mechanism of drone clusters, identify vulnerable nodes and determine the timing of intervention, resulting in poor reverse control effect.

Method used

The probability distribution difference measurement method based on discretized divergence is used to estimate the spatial distribution of the drone cluster, identify emergencies, and judge the intervention effect through the reverse control model, and determine the communication node with the highest connectivity to apply interference to achieve reverse control.

Benefits of technology

It realizes effective identification of the emergence behavior of the drone cluster and accurate selection of vulnerable nodes, ensuring the effectiveness of reverse control and preventing the disordered state and disintegration of the drone cluster.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119088078A_ABST
    Figure CN119088078A_ABST
Patent Text Reader

Abstract

The invention discloses an unmanned aerial vehicle cluster emergence behavior identification method and system, and relates to the unmanned aerial vehicle cluster control technology, and the method comprises the steps: initializing the position and speed of an unmanned aerial vehicle in an unmanned aerial vehicle cluster, and obtaining the motion state information of the unmanned aerial vehicle cluster; estimating the spatial distribution of the unmanned aerial vehicle cluster based on a discretization divergence probability distribution difference measurement method to identify the emergence behavior of the unmanned aerial vehicle cluster; and carrying out reverse control on the sub-cluster with the emergence behavior, and judging whether the reverse control is completed or not according to a failure judgment model of the unmanned aerial vehicle cluster. According to the method provided by the embodiment of the invention, modeling, observation, calculation, prediction and control of the emergence behavior of the unmanned aerial vehicle cluster are completed, effective identification of the emergence behavior of the unmanned aerial vehicle cluster and accurate selection of vulnerable nodes are realized, and reverse control is performed on the unmanned aerial vehicle cluster.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of drone cluster control technology, and in particular to a method and system for identifying emergent behavior of drone clusters. Background Art

[0002] The emergent behavior of drone swarms refers to the fact that a large number of drones establish connections with other individuals through local rules and simple interactions, and exhibit complex and organized behavior patterns as a whole. These behaviors are not obvious on a single drone, but when a large number of drones interact with each other, higher-level structures and functions will be generated. At present, there is an urgent need to reveal the operating rules and mechanisms hidden in drone swarms, identify vulnerable nodes and intervention opportunities, and support the construction of effective means of reverse control. Summary of the invention

[0003] The embodiments of the present application provide a method and system for identifying emergent behaviors of drone clusters, and propose a method for effectively identifying emergent behaviors of drone clusters and accurately selecting vulnerable nodes, so as to complete the modeling, observation, calculation, prediction and control of emergent behaviors of drone clusters, realize the effective identification of emergent behaviors of drone clusters and accurate selection of vulnerable nodes, and perform reverse control of drone clusters.

[0004] The present application embodiment provides a method for identifying emergent behavior of a drone cluster, including:

[0005] Initialize the position and speed of the drones in the drone cluster, and obtain the motion status information of the drone cluster;

[0006] Based on the probability distribution difference measurement method of discretized divergence, the spatial distribution of drone clusters is estimated to identify the emergent behavior of drone clusters;

[0007] For the sub-clusters that exhibit emergent behaviors, reverse control is performed, and whether the reverse control is completed is determined based on the failure judgment model of the drone cluster.

[0008] Optionally, the spatial distribution of the drone cluster can be estimated by measuring the difference in probability distribution based on the discretization divergence, including:

[0009] The probability distribution based on discretized divergence is used to compare the spatial distribution of drone clusters, satisfying:

[0010]

[0011] Among them, P(x) and Q(x) represent two probability distributions on the probability space Ω, p(x) and q(x) represent the probability densities of P(x) and Q(x), satisfying dP(x) = p(x)dμ(x) and dQ(x) = q(x)dμ(x), μ represents the reference distribution, p i and qi represents the probability distribution set P = {P1, P2, ..., P N} and Q = {Q1, Q2, ..., Q N}Probability value in probability space Ω.

[0012] Optionally, the spatial distribution of the drone cluster can be estimated based on the probability distribution difference measurement method of the discretization divergence, and the following method can be used to estimate the spatial distribution of the drone cluster:

[0013] For drone swarms, a nonparametric multivariate kernel density estimation method is used to estimate the spatial distribution of drone swarms in probability space at the required time. and satisfy:

[0014]

[0015] in, is the probability density function to be estimated, X1, X2, L, X N is a sample of a d-dimensional random variable, h is the smoothing coefficient, N is the sample size, and K(·) is the kernel function, that is,

[0016] The probability space is discretized and divided into multiple sub-regions so that the sub-regions are converted into corresponding probability values ​​to calculate the joint probability distribution of the probability space.

[0017] Optionally, the spatial distribution of the drone cluster can be estimated based on the probability distribution difference measurement method of the discretization divergence, and the following method can be used to estimate the spatial distribution of the drone cluster:

[0018] Based on the sliding time window, the motion state information of each UAV is continuously obtained to generate observation samples with timestamps. The spatial distribution of clusters at different times is estimated based on the observation samples.

[0019] Optionally, for the sub-cluster that exhibits emergent behavior, reverse control includes:

[0020] The communication node with the highest connectivity is determined and interference is applied to achieve reverse control of the emergent behavior of the cluster.

[0021] Optionally, judging whether the reverse control is completed according to the failure judgment model of the drone cluster includes:

[0022] The drone cluster is regarded as a network, and any drone is regarded as a node in the network. The communication between drones is represented by E ij , the drone cluster is represented as V = {V1, V2, ..., V N}, establish the failure judgment model of UAV cluster under interference conditions to meet the following requirements:

[0023]

[0024] in, u′ i represents the cluster control vector of the i-th uninterrupted UAV.

[0025] An embodiment of the present application also proposes a device for identifying emergent behavior of a drone swarm, including a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the method for identifying emergent behavior of a drone swarm as described above are implemented.

[0026] The embodiment of the present application proposes a method for effectively identifying the emergent behavior of a drone cluster and accurately selecting vulnerable nodes, which completes the modeling, observation, calculation, prediction and control of the emergent behavior of a drone cluster, realizes the effective identification of the emergent behavior of a drone cluster and accurately selects vulnerable nodes, and performs reverse control of the drone cluster.

[0027] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0029] Figure 1 The following is a flow chart of a method for identifying emergent behavior of a drone cluster according to an embodiment of the present application;

[0030] Figure 2 The initial state of the drone cluster in the application example of the method for identifying emergent behavior of a drone cluster according to an embodiment of the present application;

[0031] Figure 3 The drone interaction domain of the application example of the drone cluster emergent behavior recognition method of the embodiment of the present application;

[0032] FIG4 is a diagram showing a state where the speed synchronization of a drone cluster is completed in an application example of the method for identifying emergent behavior of a drone cluster according to an embodiment of the present application, wherein 4a is a drone before reverse control, and 4b is a drone after reverse control;

[0033] FIG5 is a diagram showing the effect of pre-reverse control of the emergent behavior of a drone swarm in an application example of the method for identifying emergent behavior of a drone swarm in an embodiment of the present application, wherein 5a is an f-divergence curve, 5b is a heading consistency change curve, 5c is an average relative distance curve, and 5d is a schematic diagram of a two-dimensional scene simulation;

[0034] Figure 6 is a diagram showing the hysteresis reverse control effect of the emergent behavior of a drone swarm in an application example of the method for identifying emergent behavior of a drone swarm according to an embodiment of the present application, wherein 6a is a heading consistency change curve, 6b is an average relative distance curve, and 6c is a schematic diagram of a two-dimensional scene simulation. DETAILED DESCRIPTION

[0035] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0036] The present application embodiment provides a method for identifying emergent behavior of a drone cluster, such as Figure 1 As shown, including:

[0037] In step S101, the positions and velocities of the drones in the drone cluster are initialized, and the motion state information of the drone cluster is obtained.

[0038] In step S102, based on the probability distribution difference measurement method of discretized divergence, the spatial distribution of the drone cluster is estimated to identify the emergent behavior of the drone cluster. For example, in some examples, Figure 2 The figure shows the mechanism for generating emergent behavior of drone clusters. In the embodiment of the present application, a measurement method for emergent behavior is established, and the spatial distribution of drone clusters is estimated.

[0039] In step S102, reverse control is performed on the sub-cluster that exhibits emergent behavior, and whether the reverse control is completed is determined based on the failure judgment model of the drone cluster.

[0040] In some embodiments, estimating the spatial distribution of the drone cluster based on the probability distribution difference measurement method of discretized divergence includes:

[0041] The probability distribution based on discretized divergence is used to compare the spatial distribution of drone clusters and analyze the emergent behavior of drone clusters from a macro level to meet the following requirements:

[0042]

[0043] Among them, P(x) and Q(x) represent two probability distributions on the probability space Ω, p(x) and q(x) represent the probability densities of P(x) and Q(x), satisfying dP(x) = p(x)dμ(x) and dQ(x) = q(x)dμ(x), μ represents the reference distribution, p i and q i represents the probability distribution set P = {P1, P2, ..., P N} and Q = {Q1, Q2, ..., Q N}Probability value in probability space Ω.

[0044] For drone swarms, the key to applying f-divergence to emergent behavior metrics is to estimate its spatial distribution in probability space at a specific time. and In some embodiments, estimating the spatial distribution of the drone cluster based on the probability distribution difference measurement method of discretized divergence further includes:

[0045] For drone swarms, a nonparametric multivariate kernel density estimation method is used to estimate the spatial distribution of drone swarms in probability space at the required time. and satisfy:

[0046]

[0047] in, is the probability density function to be estimated, X1, X2, L, X N is a sample of a d-dimensional random variable, h is the smoothing coefficient, N is the sample size, and K(·) is the kernel function, that is,

[0048] The probability space is discretized and divided into multiple sub-regions so that the sub-regions are converted into corresponding probability values ​​to calculate the joint probability distribution of the probability space.

[0049] In some embodiments, estimating the spatial distribution of the drone cluster based on the probability distribution difference measurement method of discretized divergence further includes:

[0050] Based on the sliding time window, the motion state information of each UAV is continuously obtained, and observation samples with timestamps are generated. The spatial distribution of clusters at different times is estimated based on the observation samples. In a specific example, Figure 3 As shown in the figure, when identifying the emergent behavior of a drone cluster, it is necessary to continuously obtain the motion state information of each drone and generate observation samples with timestamps. Based on the sample data, the spatial distribution of the cluster at different times can be estimated.

[0051] Drone swarms are heavily dependent on information transmission systems. Through research and analysis of swarm communication networks, in some embodiments, reverse control is performed on sub-clusters that exhibit emergent behavior, including: determining the communication nodes with the highest connectivity, achieving precise screening of vulnerable nodes, and applying interference to achieve reverse control of the cluster's emergent behavior.

[0052] In some embodiments, judging whether the reverse control is completed according to the failure judgment model of the drone cluster includes:

[0053] The drone cluster is regarded as a network, and any drone is regarded as a node in the network. The communication between drones is represented by E ij , the drone cluster is represented as V = {V1, V2, ..., V N}, establish the failure judgment model of UAV cluster under interference conditions to meet the following requirements:

[0054]

[0055] in, u′ i represents the cluster control vector of the ith undisturbed UAV. In this case, the undisturbed UAV can only exchange position and velocity information with other undisturbed cluster UAVs. As the number of exchanged information nodes decreases, the possibility of achieving coordinated motion with other UAVs in the cluster also decreases accordingly.

[0056] Figure 4 is a schematic diagram of the principle of countering emergent behavior of drone clusters. When the communication receiver of a drone is interfered, the drone will not be able to receive the motion status information of adjacent drones and will not be able to cooperate with adjacent drones in motion, such as gathering, dispersing and aligning. When the communication network of the entire drone cluster is completely suppressed, the navigation control within the drone cluster will be in a disordered state and the drone cluster will disintegrate.

[0057] In the launch phase of the drone swarm, the swarm has not yet formed and most drones are in a disordered state. If the defender can understand the situation at this stage, it can try to implement reverse control on the threatening drone swarm in advance to avoid the occurrence of swarm emergence behavior. Figure 5 shows the effect of the simulation experiment for this scenario. When the communication interference is set to 5s, the Hellinger divergence value remains basically stable, the heading consistency value is around 0.063, and the average distance value is getting larger and larger. In theory, this reverse control effect is better, but at this stage, the swarm drones are more dispersed in space, and more interference resources need to be dispatched to achieve the ideal reverse control effect.

[0058] If the reverse control measures are taken late, the cluster is in a relatively stable flight formation. Assuming that each drone still flies to the mission area when the communication is blocked, the heading, speed and relative position of each drone in this case should be basically consistent after the cluster reaches a steady state, and the cluster drones no longer need to exchange more motion state information. At this time, the reverse control of the emergent behavior of the cluster is invalid, as shown in Figure 6, the delayed reverse control effect of the emergent behavior of the drone cluster.

[0059] The embodiment of the present application proposes a method for effectively identifying the emergent behavior of a drone cluster and accurately selecting vulnerable nodes, which completes the modeling, observation, calculation, prediction and control of the emergent behavior of a drone cluster, realizes the effective identification of the emergent behavior of a drone cluster and accurately selects vulnerable nodes, and performs reverse control of the drone cluster.

[0060] The method of this application uses two indicators to reflect the order or disorder of the group state before and after reverse control based on the nature of the self-organizing process. Based on the heuristic rules followed by the cluster control algorithm, the reverse control principle of cluster emergent behavior is proposed. A judgment model for the failure of drone cluster control under reverse control conditions is constructed, and the stability of the system is analyzed. The modeling, observation, calculation, prediction and control of the emergent behavior of drone clusters are completed, and the effective identification of the emergent behavior of drone clusters and the precise selection of vulnerable nodes are achieved, laying a theoretical and technical foundation for the reverse control of drone clusters.

[0061] An embodiment of the present application also proposes a device for identifying emergent behavior of a drone swarm, including a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the method for identifying emergent behavior of a drone swarm as described above are implemented.

[0062] It should be noted that in the various embodiments of the present application, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0063] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0064] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0065] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A method for identifying emergent behavior of drone clusters, characterized in that: include: Initialize the position and speed of the drones in the drone cluster, and obtain the motion status information of the drone cluster; Based on the probability distribution difference measurement method of discretized divergence, the spatial distribution of drone clusters is estimated to identify the emergent behavior of drone clusters; For the sub-clusters that exhibit emergent behaviors, reverse control is performed, and whether the reverse control is completed is determined based on the failure judgment model of the drone cluster.

2. The method for identifying emergent behavior of drone clusters according to claim 1, characterized in that: Based on the probability distribution difference measurement method of discretization divergence, the spatial distribution of drone clusters is estimated including: The probability distribution based on discretized divergence is used to compare the spatial distribution of drone clusters, satisfying: Among them, P(x) and Q(x) represent two probability distributions on the probability space Ω, p(x) and q(x) represent the probability densities of P(x) and Q(x), satisfying dP(x) = p(x)dμ(x) and dQ(x) = q(x)dμ(x), μ represents the reference distribution, p i and q i represents the probability distribution set P = {P1, P2, ..., P N } and Q = {Q1, Q2, ..., Q N }Probability value in probability space Ω.

3. The method for identifying emergent behavior of drone clusters according to claim 2, characterized in that: Based on the probability distribution difference measurement method of discretization divergence, the spatial distribution of drone clusters is estimated, which also includes: For drone swarms, a nonparametric multivariate kernel density estimation method is used to estimate the spatial distribution of drone swarms in probability space at the required time. and satisfy: in, is the probability density function to be estimated, X1, X2, L, X N is a sample of a d-dimensional random variable, h is the smoothing coefficient, N is the sample size, and K(·) is the kernel function, that is, The probability space is discretized and divided into multiple sub-regions so that the sub-regions are converted into corresponding probability values ​​to calculate the joint probability distribution of the probability space.

4. The method for identifying emergent behavior of drone clusters according to claim 3, characterized in that: Based on the probability distribution difference measurement method of discretization divergence, the spatial distribution of drone clusters is estimated, which also includes: Based on the sliding time window, the motion state information of each UAV is continuously obtained to generate observation samples with timestamps. The spatial distribution of clusters at different times is estimated based on the observation samples.

5. The method for identifying emergent behavior of drone clusters according to claim 1, characterized in that: For sub-clusters that exhibit emergent behavior, reverse control includes: The communication node with the highest connectivity is determined and interference is applied to achieve reverse control of the emergent behavior of the cluster.

6. The method for identifying emergent behavior of drone clusters according to claim 5, characterized in that: Judging whether the reverse control is completed according to the failure judgment model of the drone cluster includes: The drone cluster is regarded as a network, and any drone is regarded as a node in the network. The communication between drones is represented by E ij , the drone cluster is represented as V = {V1, V2, ..., V N }, establish the failure judgment model of UAV cluster under interference conditions to meet the following requirements: in, u′ i represents the cluster control vector of the i-th uninterrupted UAV.

7. A drone swarm emergence behavior recognition device, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the method for identifying emergent behavior of a drone cluster as described in any one of claims 1 to 6 are implemented.