Method, system and device for reconfiguring role allocation of unmanned swarm organization and storage medium

Abstract

The application discloses a method, system and device for role allocation of unmanned cluster organization reconstruction and a storage medium. The method comprises the following steps: acquiring a plurality of basic models, and constructing a model of cluster organization reconstruction based on the plurality of basic models; establishing an index for analyzing dynamic reconstruction of roles of the model of cluster organization reconstruction; calculating trust degrees between nodes in the model of cluster organization reconstruction, and setting a model reconstruction method according to the trust degrees and the index; performing model reconstruction on the model of cluster organization reconstruction according to the model reconstruction method, and obtaining a plurality of model reconstruction solution sets; and acquiring an optimal solution set from the plurality of model reconstruction solution sets, and taking the optimal solution set as a scheme for role allocation of unmanned cluster organization reconstruction. The application has strong adaptability and improves the combat effectiveness of the whole system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned cluster organization reconstruction, in particular to a method, system and device for role allocation in unmanned cluster organization reconstruction and a storage medium. BACKGROUND

[0002] An unmanned cluster is a mobile multi-agent system with self-organizing characteristics based on a sympathetic network, which is composed of a certain number of unmanned combat equipment. In the case that the original cluster cannot normally perform combat tasks, the unmanned equipment platform can join a new cluster or undertake new tasks in the original cluster. At the cluster level, the equipment system characteristics require support for reconfigurable grouping.

[0003] In the prior art, it is generally assumed that unmanned combat equipment is in a relatively static environment, but in actual combat environment, disturbance factors exist, including external real-time variable environmental situation disturbance and internal component failure communication interruption and other disturbance factors. Disturbance can cause problems such as unmanned cluster network node failure and link interruption, which in turn leads to changes in cluster network topology and affects the combat effectiveness of the entire system. SUMMARY

[0004] The present application aims to at least solve one of the technical problems existing in the prior art. To this end, the present application provides a method, system and device for role allocation in unmanned cluster organization reconstruction and a storage medium, which can have strong adaptability and improve the combat effectiveness of the entire system.

[0005] In a first aspect, the present application provides a method for role allocation in unmanned cluster organization reconstruction, which comprises:

[0006] Obtaining a plurality of basic patterns and constructing a pattern model for cluster organization reconstruction based on the plurality of basic patterns;

[0007] Establishing an index for analyzing the dynamic reconstruction of roles in the pattern model for cluster organization reconstruction;

[0008] Calculating the trust degree between nodes in the pattern model for cluster organization reconstruction, and setting a pattern reconstruction method according to the trust degree and the index;

[0009] Reconstructing the pattern model for cluster organization reconstruction according to the pattern reconstruction method to obtain a plurality of pattern reconstruction solution sets;

[0010] Obtaining the optimal solution set from the plurality of pattern reconstruction solution sets, and taking the optimal solution set as a scheme for role allocation in unmanned cluster organization reconstruction.

[0011] Compared with the prior art, the first aspect of the present application has the following beneficial effects:

[0012] The method first acquires a plurality of basic patterns, and the pattern model of the cluster organization reconstruction is constructed by the basic patterns with actual combat significance, so that the pattern model can be more in line with the actual situation; by establishing an index of role dynamic reconstruction of the pattern model of the cluster organization reconstruction, the trust degree between nodes in the pattern model of the cluster organization reconstruction is calculated, and according to the trust degree and the index, a pattern reconstruction method is set, so that the designed pattern reconstruction method can effectively face unexpected situations occurring in the execution process, and has strong adaptability; by reconstructing the pattern model of the cluster organization reconstruction according to the pattern reconstruction method, a plurality of pattern reconstruction solution sets are obtained, and the optimal solution set is obtained from the plurality of pattern reconstruction solution sets, and the optimal solution set is used as a scheme of role allocation of the unmanned cluster organization reconstruction, and the optimal solution set is selected, so that the combat effectiveness of the whole system can be improved.

[0013] According to some embodiments of the present application, the acquiring a plurality of basic patterns comprises:

[0014] The role updating pattern, the role allocation pattern, the role evaluation pattern and the role generation pattern are acquired, wherein the role updating pattern is a new configuration target list updated by a cluster leader platform according to a switching rule; the role allocation pattern is a role allocation by the cluster leader platform according to a switching rule; the role evaluation pattern is an evaluation of each candidate scheme by the cluster leader platform; and the role generation pattern is a new role allocation scheme selected by the cluster leader platform and issued to subordinate platforms.

[0015] According to some embodiments of the present application, the index comprises:

[0016] The importance of the pattern is analyzed from the similarity of the structure, the importance of the pattern is quantified from the statistical point of view, and the current node is allocated to the processing machine that enables the current node to start earliest.

[0017] According to some embodiments of the present application, the calculating the trust degree between nodes in the pattern model of the cluster organization reconstruction comprises:

[0018] The in-degree trust degree and the out-degree trust degree are calculated by the jaccard similarity coefficient:

[0019]

[0020]

[0021] Wherein, || represents the number of elements in the symbol set, represents that the node i and the node j have similar in-degree trust degrees, represents that the node i and the node j have similar out-degree trust degrees;

[0022] normalizing the in-degree trust degree and the out-degree trust degree:

[0023]

[0024]

[0025] wherein, denotes the normalized in-degree trust degree, denotes the normalized out-degree trust degree, denotes the mean of the in-degree trust degree of all archetypes, denotes the mean of the out-degree trust degree of all archetypes, denotes the standard deviation of the in-degree trust degree of all archetypes, denotes the standard deviation of the out-degree trust degree of all archetypes;

[0026] weighting average of the normalized in-degree trust degree and the normalized out-degree trust degree, to obtain the trust degree between node i and node j in the cluster organization reconstructed archetype model:

[0027]

[0028] wherein, a denotes a weighting coefficient, 0 < a < 1.

[0029] According to some embodiments of the present application, the setting of the archetype reconstruction method according to the trust degree and the index comprises:

[0030] judging whether there is a new task requirement in the cluster organization reconstructed archetype model, and if there is a new task requirement, updating a task list;

[0031] adopting a nonlinear mixed Boolean programming method to process a node interruption scenario in the new task requirement; wherein, the nonlinear mixed Boolean programming method is a method of adopting a nonlinear programming method and adopting a Boolean mixed method;

[0032] after processing the node interruption scenario, selecting an archetype for constructing the cluster organization reconstructed archetype model and establishing an index for analyzing the cluster organization reconstructed archetype model;

[0033] calculating the trust degree between nodes in the cluster organization reconstructed archetype model, and selecting nodes with a trust degree higher than a preset value for archetype reconstruction.

[0034] According to some embodiments of the present application, the node interruption scenario comprises: there is an uncertain factor leading to that the original archetype structure does not meet the cluster requirement and there is a phenomenon of node interruption disturbing the basic archetype structure.

[0035] According to some embodiments of the present application, the optimal solution set is obtained from the plurality of solution sets of the reconstruction of the schema, comprising:

[0036] The schema reconstruction method is optimized based on a non-dominated sorting genetic method with an elitist strategy, wherein a new crowding distance calculation formula is added to the non-dominated sorting genetic method with the elitist strategy, and the optimized schema reconstruction method is obtained, wherein the new crowding distance calculation formula is:

[0037]

[0038] wherein M represents the total number of objective functions, LON[Z i ] represents the crowding distance of individual Z i , max(Z i+1 ) represents the maximum value of the population under the m objective function, min(Z M ) represents the minimum value of the population under the m objective function, g(Z i+1 ) i-1 g(Z M ) represents the objective function of individual Z i-1 ;

[0039] The optimal solution set is obtained from the plurality of solution sets of the reconstruction of the schema by using the optimized schema reconstruction method.

[0040] In a second aspect, the embodiments of the present application also provide a system for assigning roles in reconstruction of unmanned cluster organization, comprising:

[0041] A model construction unit is configured to obtain a plurality of basic schemas and construct a schema model for reconstruction of cluster organization based on the plurality of basic schemas;

[0042] An index establishment unit is configured to establish an index for dynamic reconstruction of roles in the schema model for reconstruction of cluster organization;

[0043] A trust degree calculation unit is configured to calculate trust degrees between nodes in the schema model for reconstruction of cluster organization, and set a schema reconstruction method according to the trust degrees and the index;

[0044] A schema reconstruction unit is configured to reconstruct the schema model for reconstruction of cluster organization according to the schema reconstruction method, and obtain a plurality of solution sets of reconstruction of the schema;

[0045] An optimal solution set obtaining unit is configured to obtain an optimal solution set from the plurality of solution sets of reconstruction of the schema, and take the optimal solution set as a scheme for assigning roles in reconstruction of unmanned cluster organization.

[0046] In a third aspect, an embodiment of the present application further provides an unmanned cluster organization reconstruction role allocation device, comprising at least one control processor and a memory connected with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the unmanned cluster organization reconstruction role allocation method.

[0047] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium, which stores computer executable instructions for enabling a computer to execute the unmanned cluster organization reconstruction role allocation method.

[0048] It can be understood that the beneficial effects of the above-mentioned second aspect to fourth aspect compared with the related art are the same as the beneficial effects of the above-mentioned first aspect compared with the related art, and reference can be made to the related description in the first aspect, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0049] The above and / or additional aspects and advantages of the present application will become apparent and be readily understood from the following description, taken in conjunction with the following drawings, in which:

[0050] Figure 1 is a flowchart of an unmanned cluster organization reconstruction role allocation method according to an embodiment of the present application;

[0051] Figure 2 is a schematic diagram of a morphospace evolution according to an embodiment of the present application;

[0052] Figure 3 is a schematic diagram of a morphospace variant reconstruction according to an embodiment of the present application;

[0053] Figure 4 is a flowchart of a morphospace reconstruction method according to an embodiment of the present application;

[0054] Figure 5 is a schematic diagram of a node dynamic adjustment according to an embodiment of the present application;

[0055] Figure 6 is a schematic diagram of a cluster performance unit and ARD execution task cost according to an embodiment of the present application;

[0056] Figure 7 is a schematic diagram of a cluster role reconfiguration mechanism scheme according to an embodiment of the present application;

[0057] Figure 8 is a task planning Gantt chart according to an embodiment of the present application;

[0058] Figure 9is a structural diagram of an unmanned cluster organization reconfiguration role allocation system according to an embodiment of the present application. DETAILED DESCRIPTION

[0059] Embodiments of the present application are described below in detail, examples of which are shown in the accompanying drawings, in which the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the accompanying drawings are exemplary and are for the purpose of explanation only, and are not to be understood as limiting the present application.

[0060] In the description of the present application, if there is a description to first, second, etc., it is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features or the sequence of indicated technical features.

[0061] In the description of the present application, it is to be understood that the orientation description, such as up, down, etc., is based on the orientation or positional relationship shown in the drawings, only for the purpose of describing the present application and simplifying the description, and is not to be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0062] In the description of the present application, it is to be understood that, unless otherwise explicitly limited, the words such as setting, installing, connecting, etc. should be broadly understood, and those skilled in the art can reasonably determine the specific meaning of the above words in the present application in combination with the specific content of the technical solution.

[0063] An unmanned cluster is a mobility multi-agent system composed of a certain number of unmanned combat equipment, based on a sympathetic network, so that the whole has self-organizing characteristics. Unmanned equipment platforms can join new clusters or undertake new tasks in the original cluster in the case that the original cluster cannot normally perform combat tasks. At the cluster level, the equipment system characteristic requirement is to support reconfigurable marshalling.

[0064] In the prior art, it is generally assumed that unmanned combat equipment is in a relatively static environment, but in actual combat environment, disturbance factors exist, including external real-time variable environmental situation disturbance and internal component failure communication interruption and other disturbance factors. Disturbance can cause problems such as failure of unmanned cluster network nodes, link interruption, etc., which in turn leads to changes in cluster network topology structure, affecting the combat effectiveness of the whole system.

[0065] To solve the above problems, the application first acquires a plurality of basic patterns, and a pattern model of cluster organization reconstruction constructed by basic patterns with actual combat significance can be more in line with actual conditions; by establishing an index for analyzing the role dynamic reconstruction of the pattern model of cluster organization reconstruction, the trust degree between nodes in the pattern model of cluster organization reconstruction is calculated, and according to the trust degree and the index, a pattern reconstruction method is set, so that the designed pattern reconstruction method can effectively face unexpected situations occurring in the execution process and has strong adaptability; by reconstructing the pattern model of cluster organization reconstruction according to the pattern reconstruction method, a plurality of pattern reconstruction solution sets are obtained, the optimal solution set is obtained from the plurality of pattern reconstruction solution sets, and the optimal solution set is taken as the scheme of the role allocation of the unmanned cluster organization reconstruction, so that the optimal solution set can improve the combat effectiveness of the whole system.

[0066] Reference Figure 1 The embodiment of the application provides an unmanned cluster organization reconstruction role allocation method, and the unmanned cluster organization reconstruction role allocation method comprises but is not limited to steps S100 to S500:

[0067] Step S100, a plurality of basic patterns are acquired, and a pattern model of cluster organization reconstruction is constructed based on the plurality of basic patterns;

[0068] Step S200, an index for analyzing the role dynamic reconstruction of the pattern model of cluster organization reconstruction is established;

[0069] Step S300, the trust degree between nodes in the pattern model of cluster organization reconstruction is calculated, and according to the trust degree and the index, a pattern reconstruction method is set;

[0070] Step S400, the pattern model of cluster organization reconstruction is reconstructed according to the pattern reconstruction method, and a plurality of pattern reconstruction solution sets are obtained;

[0071] Step S500, the optimal solution set is obtained from the plurality of pattern reconstruction solution sets, and the optimal solution set is taken as the scheme of the role allocation of the unmanned cluster organization reconstruction.

[0072] In steps S100 to S500 of some embodiments, in order to make the constructed cluster organization reconstructed body model more in line with the actual situation, a plurality of basic bodies are obtained, and a cluster organization reconstructed body model is constructed based on the plurality of basic bodies; in order to make the designed body reconstruction method be able to effectively face unexpected situations occurring in the execution task process, and have strong adaptability, an index for analyzing the role dynamic reconstruction of the cluster organization reconstructed body model is established, the trust degree between nodes in the cluster organization reconstructed body model is calculated, and the body reconstruction method is set according to the trust degree and the index; in order to improve the combat effectiveness of the whole system, the body reconstruction method is used to perform body reconstruction on the cluster organization reconstructed body model, a plurality of body reconstruction solution sets are obtained, an optimal solution set is obtained from the plurality of body reconstruction solution sets, and the optimal solution set is used as a scheme for distributing roles of the unmanned cluster organization reconstruction.

[0073] In some embodiments, the plurality of basic bodies are obtained by:

[0074] The role updating body, the role allocation body, the role evaluation body and the role generation body are obtained, wherein the role updating body is used for the cluster leader platform to update a new configuration target list according to a switching rule; the role allocation body is used for the cluster leader platform to allocate a candidate mode according to the switching rule; the role evaluation body is used for the cluster leader platform to evaluate each candidate scheme; and the role generation body is used for the cluster leader platform to select a new role allocation scheme and issue the new role allocation scheme to subordinate platforms.

[0075] In the embodiment, the role updating body, the role allocation body, the role evaluation body and the role generation body are all basic simple, frequently occurring and basic bodies with actual combat significance.

[0076] In some embodiments, the index includes:

[0077] The importance of the body is analyzed from the similarity of the structure, the importance of the body is quantified from the statistical point of view, and the current node is allocated to the processing machine that enables the current node to start earliest.

[0078] In the embodiment, the effect of the body reconstruction change is measured by the three constructed indexes.

[0079] In some embodiments, the trust degree between nodes in the cluster organization reconstructed body model is calculated by:

[0080] The in-degree trust degree and the out-degree trust degree are calculated by a jaccard similarity coefficient:

[0081]

[0082]

[0083] Wherein, || represents the number of elements in the set within the symbol, represents that the node i and the node j have similar in-degree trust degree, represents that the node i and the node j have similar out-degree trust degree;

[0084] Standardize the in-degree trust degree and the out-degree trust degree:

[0085]

[0086]

[0087] Wherein, represents the standardized in-degree trust degree, represents the standardized out-degree trust degree, represents the mean of the in-degree trust degree of all motifs, represents the mean of the out-degree trust degree of all motifs, represents the standard deviation of the in-degree trust degree of all motifs, represents the standard deviation of the out-degree trust degree of all motifs;

[0088] Weighted average the standardized in-degree trust degree and the standardized out-degree trust degree, to obtain the trust degree between the node i and the node j in the motif model of the cluster organization reconstruction:

[0089]

[0090] Wherein, α represents a weighting coefficient, 0<α<1.

[0091] In some embodiments, according to the trust degree and the index, the motif reconstruction method is set, comprising:

[0092] Judge whether there is a new task requirement in the motif model of the cluster organization reconstruction, if there is a new task requirement, update the task list;

[0093] Use a nonlinear mixed Boolean programming method to process the node interruption scene in the new task requirement; wherein, the nonlinear mixed Boolean programming method is a method of using nonlinear programming and using Boolean mixing;

[0094] After processing the node interruption scene, select the motif used to construct the motif model of the cluster organization reconstruction and establish the index for analyzing the motif model of the cluster organization reconstruction;

[0095] Calculate the trust degree between the nodes in the motif model of the cluster organization reconstruction, and select the nodes with the trust degree higher than a preset value for motif reconstruction.

[0096] In the embodiment, in actual combat environment, disturbance factors exist, including external real-time variable environment situation disturbance and internal component fault communication interruption and other disturbance factors. Disturbance can cause unmanned cluster network node failure, link interruption and other problems, thereby causing cluster network topology structure to change, and affecting combat effectiveness of the whole system. Therefore, in the embodiment, the nonlinear mixed Boolean programming method is used to process the node interruption scene in the new task requirement, and the node with a trust degree higher than a preset value is selected to reconstruct the module, so that the set module reconstruction method can effectively face unexpected situations in the task execution process, and has strong adaptability. Thus, the problem that disturbance can cause unmanned cluster network node failure, link interruption and other problems is solved, and the combat effectiveness of the whole system is improved.

[0097] In some embodiments, the node interruption scene includes: there are uncertain factors to cause the original module structure to not meet the cluster requirements and the phenomenon of node interruption to disturb the basic module structure.

[0098] In some embodiments, the optimal solution set is obtained from a plurality of module reconstruction solutions, including:

[0099] Based on the non-dominated sorting genetic method with an elite strategy, a new crowding distance calculation formula is added to the non-dominated sorting genetic method with an elite strategy to optimize the module reconstruction method, and an optimized module reconstruction method is obtained, wherein the new crowding distance calculation formula is:

[0100]

[0101] wherein M represents the total number of objective functions, LON[Z i ] represents the crowding distance of individual Z i , represents the maximum value of the population under the m objective function, represents the minimum value of the population under the m objective function, g(Z i+1 ) M represents the objective function of individual Z i+1 , g(Z i-1 ) M represents the objective function of individual Z i-1 ;

[0102] The optimal solution set is obtained from a plurality of module reconstruction solutions by the optimized module reconstruction method.

[0103] In the embodiment, the optimal solution set is obtained from a plurality of module reconstruction solutions, and the optimal solution set is used as a scheme of unmanned cluster organization reconstruction role allocation, which can further improve the combat effectiveness of the whole system.

[0104] For the convenience of those skilled in the art, a set of best embodiments is provided as follows:

[0105] I. The motif model of cluster organization reconstruction.

[0106] Motif is a sub-network structure that repeats in a specific network or different networks, which can reflect the functions that can be effectively implemented in a framework. Youngwoo Lee and Taeisk Lee published a research in 2014 that the operational measurement of operational effectiveness is a very challenging task due to the large amount of complexity presented by the rich operational environment. They proposed a new operational effectiveness measurement method based on motif. They abstracted two motifs from the operational network according to the characteristics of the operational task and the characteristics of the network motif: independent attack motif and joint attack motif. Then, Liu Jiajie of Tsinghua University proposed a function-oriented unmanned aerial vehicle cluster motif, establishing six motifs: reconnaissance motif, attack motif, reconnaissance and attack motif, reconnaissance and control attack motif, reconnaissance and control reconnaissance and attack motif, and reconnaissance and control attack evaluation motif. This embodiment introduces the motif into the cluster organization reconstruction role allocation mechanism by learning from the motif.

[0107] Firstly, according to the frequency of the motif, the basic motif is found from the cluster organization reconstruction role allocation, which is used as the basic cluster allocation structure unit. Secondly, cluster organization reconstruction is a large-scale group cooperation, and due to the scale-free nature of the motif, small-scale cluster network allocation can be designed for large-scale cluster role allocation. Finally, the motif has a simple basis, and the internal structure of the motif and the interaction structure between the motifs are simple and have fewer connections. For the communication connection between individuals restricted by geographical environment, weather and climate and many other restrictions, the motif is suitable for the design of cluster-level operations. On the other hand, the cluster access control mechanism directly allocates and authorizes from resources to roles, and does not effectively explain how the cluster roles are dynamically allocated. This embodiment abstracts four basic motifs that are simple, frequently occurring and have practical operational significance in the network from the cluster cooperative operational network, establishes a motif model of cluster organization reconstruction, and the four motif-based allocation units are as follows:

[0108] Role update motif: the cluster leader platform updates the new configuration target list according to the switching rule. According to its function, this motif is named as update motif. At the same time, in order to express conveniently, the update motif is denoted as a.

[0109] Role allocation motif: the cluster leader platform allocates the candidate mode according to the switching rule. According to its function, this motif is named as allocation motif. At the same time, in order to express conveniently, the allocation motif is denoted as b.

[0110] Role evaluation motif: the cluster leader platform evaluates each candidate scheme. According to its function, this motif is named as evaluation motif. At the same time, in order to express conveniently, the evaluation motif is denoted as c.

[0111] Role generation module: cluster leader platform selects new role allocation scheme and issues to the subordinate platforms. According to its function, this module is named as generation module. At the same time, in order to express conveniently, generation module is recorded as d.

[0112] The definition of closed four-order module subgraph and non-closed four-order module subgraph in the four-order module subgraph is established. The cluster organization reconstruction flexible mechanism is generated from the scheme of cluster leader platform, recorded as directed network G = < V, E >, where V represents the set of nodes, E represents the set of edges. E xy represents the directed edge pointed by node x to y.

[0113] Definition: closed four-order module subgraph for a given four-order module subgraph G i , if there exists a set of edges e ab , e bc , e cd , e da ∈ E, then G i is called a closed four-order module subgraph.

[0114] Definition: non-closed four-order module subgraph for a given four-order module subgraph G j , if there exists a set of edges not belonging to E, then G j is called a non-closed four-order module subgraph.

[0115] The basic structure of cluster organization reconstruction flexible mechanism is set as non-closed four-order module subgraph, and mechanism design needs to go through four steps of (1)-(4) in turn.

[0116] For a given non-closed four-order module subgraph, module {a, b, c, d} ∈ N, the connection of single module only contains four possibilities of positive, negative, bidirectional and not connected. As shown in Figure 2 , then Num(e ac ) = Num(e bd ) = Num(e da ) = 4. Thus, the connection mode of closed four-order module subgraph has 64 kinds of scheme outputs. Similarly, the non-closed four-order module subgraph has e da = null, and its connection mode has 64 kinds. Then all evolution models have 9 kinds of evolution modes with 2 9 64 kinds of connection modes. The adjacency matrix C = [C ij ] n×n of network can be expressed as:

[0117]

[0118] cij 1 if there is a connection between nodes, otherwise 0.

[0119] Because the direction determines the complexity of the edge and node, the variants of the fourth-order directed closed motif and the fourth-order directed non-closed motif are different, and need to be considered respectively, such as Figure 3 In order to measure the effect of motif reconstruction change, three indexes need to be established for analysis, and role dynamic reconfiguration mainly relies on three indexes:

[0120] 1. Analyze the similarity of the structure from the importance of the motif;

[0121] 2. Quantify the importance of the motif from the statistical point of view;

[0122] 3. Assign the node to the processing machine that starts it earliest.

[0123] The task completion time (T_mct) is set as:

[0124] T mct = max{Fin_T}

[0125] Fin_T represents the set of time when all task changes are completed.

[0126] In the fourth-order directed motif, let be the adjacency matrix of i, in order to distinguish the directionality, the in-degree and the out-degree are introduced, so the degree of node i is divided into total degree k total (i), in-degree k in (i), and out-degree k out (i), where:

[0127] k total (i) = k in (i) + k out (i)

[0128]

[0129]

[0130] Where j represents the adjacent point of i, and V(i) represents the set of j.

[0131] In the directed weighted network model instance established in the experiment, when k≤10, the in-out degree distribution of the network is a gentle process. The degree distribution function of the network whose degree distribution obeys the Poisson distribution is:

[0132] p(k) = λ k e -k / k!.

[0133] The calculation of the correlation degree of nodes is defined as follows:

[0134] represents that node i and node j have similar in-degree trust degree, represents that node i and node j have similar out-degree trust degree, represents the normalized represents the normalized T ij represents the weighted trust degree of node i and node j.

[0135] The correlation degree of nodes is defined by using the Jaccard similarity coefficient and

[0136]

[0137]

[0138] wherein || represents the number of elements in the set in the symbol, to avoid the case that the out-degree and the in-degree of the module are subtracted, the denominator is 1 after the denominator, and then normalized:

[0139]

[0140]

[0141] wherein, are the average values of the and of all modules, and are the standard deviations of all modules.

[0142] Finally, the trust degree between node i and node j is obtained by weighted average using the two similarity trust degrees.

[0143]

[0144] wherein 0<α<1 is the weighting coefficient.

[0145] II. Module reconstruction method

[0146] The cluster role reconfiguration mechanism is responsible for the execution of the cluster head platform in the cluster layer, which belongs to the task adaptive mechanism in the task domain. Its main goal is to adapt to the new state or new task requirements in the cluster, to make online adaptive planning, and to flexibly redistribute the cluster roles to support the cluster task adaptation. The starting conditions are the following two events: the superior issues a new task or finds that there is a cooperation state failure in the internal. The rule base includes task update rules, state detection rules, role allocation rules, role evaluation rules, etc. The specific steps of the mechanism execution are:

[0147] (1) The cluster head platform produces an update candidate new configuration mode according to the switching rules;

[0148] (2) The cluster head platform allocates the candidate mode according to the switching rules;

[0149] (3) The cluster head related platform evaluates the candidate mode;

[0150] (4) The cluster head platform selects a new role allocation scheme and issues it to the subordinate platforms.

[0151] The mechanism execution needs to meet the following preconditions: online role reconfiguration authorization. The mechanism execution needs to be supported by the cluster role online allocation algorithm, which can automatically generate a role allocation scheme within a short time period for the updated task list or state event, and preliminarily complete the task or cooperation compatibility judgment. This mechanism can be part of the command and control, and can be compiled into a software and hardware support cluster organization reconstruction function of the cluster head platform. The adjacency matrix considers the correlation degree of the nodes, and analyzes the structure of the directed graph from a quantitative perspective. Therefore, the four-order module can be mapped into the adjacency matrix and input into the algorithm for solution. In the adjacency matrix of the directed graph:

[0152] The meaning of the ith row: the arc with node vi as the tail (i.e. the out-degree edge).

[0153] The meaning of the ith column: the arc with node vi as the head (i.e. the in-degree edge).

[0154] Analysis 1: The adjacency matrix of the directed graph may be asymmetric;

[0155] Analysis 2: The out-degree k of the vertex out (i) is equal to the sum of the elements in the ith row, and the in-degree k of the vertex in (i) is equal to the sum of the elements in the ith column, and the degree k of the vertex total (i) is equal to the sum of the elements in the ith row plus the sum of the elements in the ith column, and the adjacency matrix of the weighted graph is represented as:

[0156]

[0157] where W ijrepresents the weight on the edge, <vi,vj>or(vi,vj)∈VR represents the edge.

[0158] The scheduling problem in the cluster is a core problem of the cluster organization reconstruction role allocation. The multi-dimensional dynamic scheduling method is often used for equipment scheduling in war and has achieved good results. Through the improvement of the multi-dimensional dynamic scheduling method, the design of the reconstruction method, the logical structure reference Figure 4 .

[0159] Therefore, in the research, the reconstruction method is used, and in the case of a given basic four-order non-closed loop module organization structure, it is judged whether there is a new task requirement, and the task module is updated. The new task requirement considers two cases:

[0160] 1. There are uncertain factors that cause the original four-order non-closed module structure to not meet the cluster requirements.

[0161] 2. There are node interruptions and other phenomena that disturb the basic four-order module structure.

[0162] Among them, the embodiment believes that the use of softening can process the node interruption of the uncertain factors and disturbances of the cluster task link, refer to Figure 5 , and a nonlinear mixed Boolean programming method is used to establish a mathematical model to optimize the risk.

[0163] The reconfigurable softening network is a variable network system that can meet the changing task requirements by adjusting the function, performance or corresponding relationship of the network nodes through reconfigurable technology. In order to meet the adaptability requirements of the battlefield, the network organization structure must be variable. The pseudo code of the specific reconstruction method is shown in Table 1.

[0164] Table 1

[0165]

[0166] The cluster role reconfiguration mechanism design problem is actually a multi-objective optimization problem, which solves the task planning optimization problem that meets three objectives. The MOEA called NSGA-2 (non-dominated sorting genetic algorithm with elitist strategy) is used, which has better performance in the same algorithm. In the process of less iteration, it has better performance.

[0167] For a multi-objective optimization problem, there may be more than one optimal solution; rather, there should be a set of optimal solutions. This set of optimal solutions is usually called a non-dominated solution set, or Pareto set, for the corresponding multi-objective optimization problem, and each solution in the set is called a Pareto solution. There are many methods for solving multi-objective optimization problems, such as common goal programming methods, goal decomposition methods, and goal reduction methods (representing multiple objectives as one). Evolutionary algorithms have a natural advantage in solving multi-objective problems. An evolutionary multi-objective optimization algorithm can optimize multiple objective functions simultaneously and output a set of non-dominated Pareto solutions, thus effectively solving multi-objective problems.

[0168] This embodiment is based on a non-dominated sorting genetic algorithm with an elitist strategy. A new crowding distance calculation formula is added to the non-dominated sorting genetic algorithm with an elitist strategy to optimize the motif reconstruction method, resulting in an optimized motif reconstruction method. The new crowding distance calculation formula is as follows:

[0169]

[0170] Where M represents the total number of objective functions, LON[Z i ] represents individual Z i Crowding distance, This represents the maximum value of the population under objective function m. Let g(Z) represent the minimum value of the population under objective function m. i+1 ) M Individual Z i+1 The objective function, g(Z) i-1 ) M Individual Z i-1 The objective function.

[0171] It should be noted that the non-dominated sorting genetic algorithm with elite strategy in this embodiment is existing technology and will not be described in detail in this embodiment.

[0172] The optimal solution set is obtained from multiple motif reconstruction solution sets using the optimized motif reconstruction method. See Table 2 for the pseudocode of the optimization process of the optimized motif reconstruction method; see Table 3 for the pseudocode of the crossover and mutation operators; and see Table 4 for the pseudocode of the optimization process for crowding distance allocation, which includes crowding degree sorting.

[0173] Table 2

[0174]

[0175] Table 3

[0176]

[0177] Table 4

[0178]

[0179] In order to better illustrate, the following experiments are carried out:

[0180] Referring to Figure 6 , knowing the current ARD execution cost, the war expert estimates the time to complete each task according to the difficulty of the task, the type and quantity of the model required to perform the task, and the battlefield situation, and the cluster role reconfiguration mechanism scheme is referred to Figure 7 .

[0181] The existing available unmanned aerial vehicle resources are 3 relay unmanned aerial vehicles, 20 decision unmanned aerial vehicles, 15 reconnaissance unmanned aerial vehicles, 18 attack unmanned aerial vehicles and 9 reconnaissance and attack integrated unmanned aerial vehicles. After analyzing the task elements such as task difficulty and enemy target quantity, quantifying the ability demand and communication demand of the task, calculating the model demand vector of the task, and estimating their completion time, referring to Figure 6 , a feasible task execution order is selected, and a dynamic list scheduling is run based on MATLAB to output the basic task start time and end time.

[0182] This order is used as input, and the main purpose is to design a disturbance point to observe whether it quickly adapts, at this time, the model reconstruction method of the embodiment is used to calculate the set ts and total under the condition of unexpected changes, and the scheduling difference is represented by a Gantt chart.

[0183] Referring to Figure 8 , in view of the disturbance point U1 and the emergence of a new attack task, U3 quickly changes to cooperate with U4 to complete the scheduling of this task. According to the experimental results, when the unmanned aerial vehicle cluster executes the task, in the face of unexpected situations such as mobilization of the organization or combat task, the model reconstruction method of the embodiment is helpful for the rapid adjustment of the unmanned aerial vehicle cluster task chain and the completion of the mission task. The simulation experiment of the embodiment uses the experimental design method to verify the cluster role reconfiguration mechanism scheme, designs an unexpected node, when the unexpected node occurs, U3 and U4 quickly adjust the task in a short time, and the effectiveness of the cluster role reconfiguration mechanism scheme is preliminarily obtained.

[0184] Referring to Figure 9 , the embodiment of the application also provides an unmanned cluster organization reconstruction role allocation system, and the unmanned cluster organization reconstruction role allocation system comprises a model construction unit 100, an index establishment unit 200, a trust degree calculation unit 300, a model reconstruction unit 400 and an optimal solution set acquisition unit 500, wherein:

[0185] The model construction unit 100 is configured to acquire a plurality of basic models, and construct a model of the cluster organization reconstruction based on the plurality of basic models.

[0186] The index establishing unit 200 is configured to establish an index for analyzing the role dynamic reconstruction of the model of the cluster organization reconstruction.

[0187] The trust degree calculation unit 300 is configured to calculate the trust degree between nodes in the model of the cluster organization reconstruction, and set a model reconstruction method according to the trust degree and the index.

[0188] The model reconstruction unit 400 is configured to perform model reconstruction on the model of the cluster organization reconstruction according to the model reconstruction method, and obtain a plurality of model reconstruction solution sets.

[0189] The optimal solution set acquisition unit 500 is configured to acquire an optimal solution set from the plurality of model reconstruction solution sets, and take the optimal solution set as a scheme of the role allocation of the cluster organization reconstruction.

[0190] It should be noted that, since the unmanned cluster organization reconstruction role allocation system in the embodiment and the unmanned cluster organization reconstruction role allocation method described above are based on the same inventive concept, the corresponding contents in the method embodiment are also applicable to the system embodiment, and will not be described in detail here.

[0191] The embodiment of the application further provides an unmanned cluster organization reconstruction role allocation device, which comprises at least one control processor and a memory in communication connection with the at least one control processor.

[0192] The memory is a non-transitory computer readable storage medium, which can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0193] The non-transitory software programs and instructions required for the unmanned cluster organization reconstruction role allocation method of the above-mentioned embodiment are stored in the memory, and when executed by the processor, the unmanned cluster organization reconstruction role allocation method of the above-mentioned embodiment is executed, for example, the method steps S100 to S500 in the above-mentioned method are executed. Figure 1

[0194] ​The system embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purposes of the embodiments.

[0195] The embodiment of the present application further provides a computer readable storage medium, which stores computer executable instructions, and the computer executable instructions are executed by one or more control processors, so that the one or more control processors execute a method embodiment of the unmanned cluster organization reconstruction role allocation method. Figure 1

[0196] Those skilled in the art can understand that all or some steps in the above disclosed method and system can be implemented as software, firmware, hardware and appropriate combinations thereof. Some or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software can be distributed on a computer readable medium, which can include computer storage media (or non-transitory media) and communication media (or transitory media). As known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. In addition, as known to those skilled in the art, communication media generally includes computer readable instructions, data structures, program modules or other data in modulated data signals such as carrier waves or other transport mechanisms, and can include any information delivery medium.

[0197] The above is a specific description of the preferred embodiments of the present application, but the embodiments of the present application are not limited to the above described embodiments. Those skilled in the art can make various equivalent modifications or replacements without departing from the spirit of the embodiments of the present application, and these equivalent modifications or replacements are all included in the scope defined by the claims of the embodiments of the present application.​

Claims

1. A method for reorganizing and assigning roles in an unmanned cluster organization, characterized in that, The unmanned cluster organization reconstruction role allocation method comprises: a plurality of basic patterns are acquired, and a pattern model of cluster organization reconstruction is constructed based on the plurality of basic patterns; an index for analyzing role dynamic reconstruction of the pattern model of cluster organization reconstruction is established; a trust degree between nodes in the pattern model of cluster organization reconstruction is calculated, and a pattern reconstruction method is set according to the trust degree and the index, wherein the index comprises similarity of an analysis structure from a pattern importance, quantification of a pattern importance from a statistical perspective, and allocation of a current node to a processing machine that enables the current node to start earliest; specifically, it is judged whether there is a new task requirement in the pattern model of cluster organization reconstruction, and if there is a new task requirement, a task list is updated; a nonlinear mixed Boolean programming method is used to process a node interruption scenario in the new task requirement; wherein the nonlinear mixed Boolean programming method is a nonlinear programming method and a Boolean mixed method; after processing the node interruption scenario, a pattern used to construct the pattern model of cluster organization reconstruction is selected, and an index for analyzing the pattern model of cluster organization reconstruction is established; a trust degree between nodes in the pattern model of cluster organization reconstruction is calculated, and a node with a trust degree higher than a preset value is selected for pattern reconstruction; the pattern model of cluster organization reconstruction is reconstructed according to the pattern reconstruction method, and a plurality of pattern reconstruction solution sets are obtained; an optimal solution set is acquired from the plurality of pattern reconstruction solution sets, and the optimal solution set is taken as an unmanned cluster organization reconstruction role allocation scheme.

2. The method of claim 1, wherein, The plurality of basic patterns are acquired, comprising: a role update pattern, a role allocation pattern, a role evaluation pattern and a role generation pattern are acquired, wherein the role update pattern is a new configuration target list updated by a cluster leader platform according to a switching rule; the role allocation pattern is a candidate mode allocated by the cluster leader platform according to the switching rule; the role evaluation pattern is an evaluation of each candidate scheme by the cluster leader platform; and the role generation pattern is a new role allocation scheme selected by the cluster leader platform and issued to subordinate platforms.

3. The method of claim 1, wherein, The trust degree between nodes in the pattern model of cluster organization reconstruction is calculated, comprising: an in-degree trust degree and an out-degree trust degree are calculated by a jaccard similarity coefficient: Wherein, || represents the number of elements in the set within the symbol, represents that the node i and the node j have similar in-degree trust degrees, represents that the node i and the node j have similar out-degree trust degrees; the in-degree trust degree and the out-degree trust degree are standardized: wherein denotes the in-degree trustworthiness normalized, denotes the out-degree trustworthiness normalized, , denotes the mean of the in-degree trustworthiness of all motifs, denotes the mean of the out-degree trustworthiness of all motifs, denotes the standard deviation of the in-degree trustworthiness of all motifs, denotes the standard deviation of the out-degree trustworthiness of all motifs; the standardized in-degree trust degree and the standardized out-degree trust degree are weighted and averaged to obtain a trust degree between node i and node j in the pattern model of cluster organization reconstruction: wherein represents a weighting coefficient, .

4. The method of claim 1, wherein, The node interruption scenario comprises: an original pattern structure does not meet cluster requirements due to uncertain factors, and a basic pattern structure is disturbed by a phenomenon of node interruption.

5. The method of claim 1, wherein, The optimal solution set is acquired from the plurality of pattern reconstruction solution sets, comprising: a non-dominated sorting genetic method with an elitist strategy is used to optimize the pattern reconstruction method by adding a new crowding distance calculation formula to the non-dominated sorting genetic method with the elitist strategy, to obtain an optimized pattern reconstruction method, wherein the new crowding distance calculation formula is: where M represents the total number of objective functions, represents the crowding distance of the individual , represents the maximum value of the population under the m objective function, represents the minimum value of the population under the m objective function, represents the objective function of the individual , represents the objective function of the individual ; An optimal solution set is obtained from the plurality of solution sets of the model body reconstruction.

6. An unmanned swarm organization reconfiguration role assignment system, characterized by, The unmanned cluster organization reconstruction role allocation system comprises: A model construction unit is configured to obtain a plurality of basic model bodies and construct a model body model of cluster organization reconstruction based on the plurality of basic model bodies; An index establishment unit is configured to establish an index for analyzing role dynamic reconstruction of the model body model of cluster organization reconstruction; A trust degree calculation unit is configured to calculate trust degrees between nodes in the model body model of cluster organization reconstruction, and set a model body reconstruction method according to the trust degrees and the index, wherein the index comprises similarity of an importance analysis structure of a model body, quantification of importance of the model body from a statistical perspective, and allocation of a current node to a processing machine that enables the current node to start earliest. It is determined whether there is a new task requirement in the model body model of cluster organization reconstruction, and if there is a new task requirement, a task list is updated. A nonlinear mixed Boolean programming method is used to process a node interruption scenario in the new task requirement; wherein the nonlinear mixed Boolean programming method is a nonlinear programming method and a Boolean mixed method. After processing the node interruption scenario, a model body used to construct the model body model of cluster organization reconstruction is selected, and an index for analyzing the model body model of cluster organization reconstruction is established. Trust degrees between nodes in the model body model of cluster organization reconstruction are calculated, and nodes with trust degrees higher than a preset value are selected for model body reconstruction. A model body reconstruction unit is configured to perform model body reconstruction on the model body model of cluster organization reconstruction according to the model body reconstruction method, and obtain a plurality of solution sets of model body reconstruction. An optimal solution set is obtained from the plurality of solution sets of model body reconstruction, and the optimal solution set is used as a scheme of unmanned cluster organization reconstruction role allocation.

7. An unmanned swarm organization reconfiguration role assignment apparatus characterized by, The computer readable storage medium stores computer executable instructions for causing a computer to execute the unmanned cluster organization reconstruction role allocation method.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer executable instructions for causing a computer to execute the unmanned cluster organization reconstruction role allocation method.

Images