A collaborative task efficiency analysis method and device based on ring network fusion

Abstract

The application discloses a kind of based on ring network fusion's collaborative task efficiency analysis method and device, it is related to efficiency analysis technical field.The unmanned aerial vehicle collaborative task system is simplified as four types of node types containing OODA ring;Based on the four types of nodes of OODA ring, determine actual connection edge, obtain task network, extract generalized OODA ring from task network, cover collaborative relationship;Generalized OODA ring node attribute quantization framework based on task demand is built, including task-driven node attribute extraction, quantization method and data acquisition path and data normalization processing;Based on generalized OODA ring node attribute quantization framework, nonlinear aggregation is carried out, and the efficiency of task network is comprehensively analyzed based on the probability of task success.The problem that the prior art cannot accurately capture the efficiency change of unmanned aerial vehicle collaborative task system in dynamic confrontation process, and is limited in practical application, is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of performance analysis, in particular to a collaborative task performance analysis method and device based on ring network fusion. BACKGROUND

[0002] In the unmanned aerial vehicle (UAV) collaborative task system, the complex network theory provides a powerful tool for analyzing the structural characteristics of the task system. With the help of this theory, the topological structure of the UAV collaborative task system can be clearly presented.

[0003] However, the analysis method based on the complex network theory mainly focuses on the static structural characteristics of the network. This results in that the analysis result can only reflect the static structural characteristics of the UAV collaborative task system at a certain moment, and cannot reflect the dynamic change process. In actual UAV collaborative task scenarios, the task system is not in a static state. The existing technology based on the complex network theory cannot accurately capture the performance changes of the UAV collaborative task system in such a dynamic confrontation process, thereby affecting the accuracy of the analysis of the task execution situation, leading to low task execution efficiency or failure to achieve the expected goal.

[0004] Currently, there is a method for analyzing the performance of OODA ring theory related tasks based on neural networks. Although this method has certain self-adaptive learning ability, it relies heavily on a large number of data samples for training. However, in actual UAV collaborative tasks, it is often difficult to obtain comprehensive and high-quality data samples. SUMMARY

[0005] In the embodiments of the present application, by providing a collaborative task performance analysis method based on ring network fusion, the problem that the prior art cannot accurately capture the performance changes of the UAV collaborative task system in the dynamic confrontation process, affecting the accuracy of the analysis of the task execution situation, leading to low task execution efficiency or failure to achieve the expected goal, and being limited in actual application is solved.

[0006] In the first aspect, the embodiments of the present application provide a collaborative task performance analysis method based on ring network fusion, which comprises: simplifying the UAV collaborative task system into four types of nodes containing OODA ring; wherein the node types include perception nodes, decision nodes, execution nodes and target nodes; determining the actual connection edge based on the four types of nodes of the OODA ring, obtaining the task network, extracting the generalized OODA ring from the task network, and covering the collaborative relationship; constructing a generalized OODA ring node attribute quantification framework based on task requirements, including task-driven node attribute extraction, quantification method and data acquisition path, and data normalization processing; based on the generalized OODA ring node attribute quantification framework, performing nonlinear aggregation, and comprehensively analyzing the task network performance based on the probability of task success.

[0007] In a possible implementation, the perception nodes are configured to acquire environment information, collect target information and monitor the environment; the decision nodes are configured to process information, analyze a scene situation, formulate a decision scheme and issue an instruction; the execution nodes are configured to perform an operation task; and the target nodes are configured to specify a key position.

[0008] In a possible implementation, the four types of nodes in the OODA loop are used to determine actual connection edges, including: when each type of node in the four types of nodes interacts with the four types of nodes including itself, considering both the information sending and receiving directions, regarding the single direction information interaction between each type of node and the four types of nodes including itself as a potential connection edge; and determining the actual connection edges based on the potential connection edges.

[0009] In a possible implementation, the generalized OODA loop is extracted from the task network to cover the cooperative relationship, including: the cooperative relationship includes cooperative perception connection edges, cooperative decision connection edges and cooperative execution connection edges; and the expression of the task network is: ; wherein, is the task network, and the nodes include perception nodes , decision nodes , execution nodes and target nodes ; the actual connection edges include cooperative perception connection edges from the perception nodes to the perception nodes, decision relationship connection edges from the perception nodes to the decision nodes, first command relationship connection edges from the decision nodes to the perception nodes, cooperative decision connection edges from the decision nodes to the decision nodes, second command relationship connection edges from the decision nodes to the execution nodes, cooperative execution connection edges from the execution nodes to the execution nodes, work relationship connection edges from the execution nodes to the target nodes and reconnaissance relationship connection edges from the target nodes to the perception nodes; a counter is initialized to 0, used to count the number of target nodes corresponding to the task loop; and starting from the target nodes , a depth-first search is performed to visit adjacent nodes, each time a new node is visited, the next recursive call is performed starting from the new node to continue the depth-first search; a node set and a node type order of a current path are recorded; and in the process of the traversal, a node type constraint is followed; the node type constraint includes: the next step starting from the target nodes is a perception node ; the next step starting from the perception nodes is a perception node or a decision node ; the next step starting from the decision nodes is a decision node or an execution node ; and the next step starting from the execution nodes The next step after starting is the execution node. or target node ; except for the target node in the path In addition, other nodes are not traversed repeatedly; when traversing to the execution node, check if there is a job relationship connection edge from the execution node to the target node. If there is, increment the counter value by 1; after each recursive call, remove the current node from the node set, explore other paths, and reduce invalid search paths through pruning and backtracking mechanisms to obtain a generalized OODA cycle.

[0010] In one possible implementation, the construction of a generalized OODA node attribute quantification framework based on task requirements includes: task-driven node attribute extraction, which includes parsing task intent, analyzing scenario requirements, and identifying constraints; combining expert experience and data-driven approaches to select core attributes for each node type related to the task; quantification methods and data acquisition paths, which include adopting a fifth-order quantification criterion and selecting appropriate quantification methods and data acquisition methods according to the type of core attributes; wherein the fifth-order quantification criterion includes experimental testing, simulation exercises, historical data statistical analysis, adversarial analysis, and comprehensive integration methods; and data normalization processing, which includes converting core attribute data of different dimensions or ranges to a unified dimension or range range.

[0011] In one possible implementation, the step-by-step attribute quantization framework based on generalized OODA performs nonlinear aggregation and conducts a comprehensive analysis of the task network performance based on the probability of task success, including: based on Obtain the probability of each node type completing the task; where, for or or or , To determine the probability that a sensing node will complete its task. The probability that a decision node will complete its task. The probability that the execution node will complete the task. The probability of completing the task for the target node. For the corresponding node's first The weight of each influencing factor For the corresponding node's first The membership degree or probability of each influencing factor in completing the task is used; based on the probability of each node type completing the task, the minimum value method is used to chain and aggregate the node sets to obtain the task completion rate of each node set, expressed as: ;in, For the first Task completion rate of a set of nodes For the first The number of nodes of a generalized OODA loop; based on the task completion rate of each node set, the multiple generalized OODA loops of the target node are regarded as a parallel structure by using a probabilistic parallel model, the probability of achieving target success is obtained, and the expression is: ; wherein, is the probability of achieving the first target success, is the task completion rate of the first node set, is the total number of node sets that can achieve the first target, the probability of task success is obtained based on the probability of achieving target success, and the task network performance is comprehensively analyzed based on the probability of task success, and the expression is: ; wherein, is the probability of task success, is the importance of achieving the target, is the probability of achieving the first target success, is the total number of targets achieved.

[0012] In a second aspect, the embodiments of the present application provide a collaborative task performance analysis device based on ring network fusion, which comprises: a simplification module for simplifying a UAV collaborative task system into four types of nodes containing OODA loops; wherein, the node types include perception nodes, decision nodes, execution nodes and target nodes; an extraction module for determining actual connection edges based on the four types of OODA loop nodes, obtaining a task network, and extracting generalized OODA loops from the task network to cover the collaborative relationship; a construction module for constructing a generalized OODA loop node attribute quantification framework based on task requirements, including task-driven node attribute extraction, quantification method and data acquisition path, and data normalization processing; and an analysis module for nonlinear aggregation based on the generalized OODA loop node attribute quantification framework, and comprehensive analysis of the task network performance based on the probability of task success.

[0013] In a third aspect, the embodiments of the present application provide a collaborative task performance analysis server based on ring network fusion, comprising a memory and a processor; the memory is used to store computer executable instructions; the processor is used to execute the computer executable instructions to realize the method of the first aspect or any possible implementation manner of the first aspect.

[0014] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, which stores executable instructions, and the computer executes the executable instructions to realize the method of the first aspect or any possible implementation manner of the first aspect.

[0015] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects: the embodiments of the present application provide a collaborative task effectiveness analysis method based on ring network fusion, which simplifies the unmanned aerial vehicle collaborative task system into four types of nodes containing the OODA ring. Based on the four types of nodes of the OODA ring, the actual connection edges are determined, the task network is obtained, the generalized OODA ring is extracted from the task network, and the collaborative relationship is covered. A generalized OODA ring node attribute quantification framework based on task requirements is constructed, including task-driven node attribute extraction, quantification method and data acquisition path, and data normalization processing. Based on the generalized OODA ring node attribute quantification framework, nonlinear aggregation is performed, and the task network effectiveness is comprehensively analyzed based on the probability of task success. The task system structure and the collaborative relationship can be accurately presented, and through node attribute quantification and nonlinear aggregation, scientific and comprehensive analysis of the task network effectiveness is realized. The problem that the prior art cannot accurately capture the effectiveness change of the unmanned aerial vehicle collaborative task system in the dynamic confrontation process, which affects the accuracy of the analysis of the task execution situation, leads to low task execution efficiency or failure to achieve the expected goal, and is limited in actual application is solved. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the description of the embodiments of the present application or the prior art will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings without creative labor based on these drawings.

[0017] Figure 1 A flowchart of a collaborative task effectiveness analysis method based on ring network fusion provided by the embodiments of the present application;

[0018] Figure 2 A schematic diagram of a collaborative task effectiveness analysis device based on ring network fusion provided by the embodiments of the present application;

[0019] Figure 3 A schematic diagram of a collaborative task effectiveness analysis server based on ring network fusion provided by the embodiments of the present application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0021] The following description of the technology involved in the embodiments of the present application is provided to facilitate understanding of the present application. It should be understood that these are merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Also, in the following description, descriptions of some well-known functions and structures are omitted for clarity and conciseness.

[0022] The embodiments of the present application provide a collaborative task effectiveness analysis method based on ring network fusion, as shown in the following formula (1) : Figure 1 The method comprises steps S101 to S104. Among them, Figure 1 The steps shown in the embodiments of the present application are only one execution order, and do not represent the only execution order of the collaborative task effectiveness analysis method based on ring network fusion. As long as the final result can be achieved, Figure 1 The steps shown in the embodiments of the present application are only one execution order, and do not represent the only execution order of the collaborative task effectiveness analysis method based on ring network fusion. As long as the final result can be achieved,

[0023] S101: Simplify the unmanned aerial vehicle collaborative task system into four types of nodes containing OODA ring. Among them, the node types include perception nodes, decision nodes, execution nodes and target nodes.

[0024] OODA ring is a decision-making process used to describe the individual or organization in response to complex, dynamic environment.

[0025] The perception node is used to obtain environmental information, collect target information and monitor the environment, including meteorological sensors, environmental monitoring unmanned aerial vehicles, traffic cameras, satellites and sensor networks.

[0026] Specifically, meteorological sensors can detect meteorological data such as temperature, humidity, wind speed and other information in various ways. Environmental monitoring unmanned aerial vehicles can monitor specific areas in a larger range, and traffic cameras can obtain real-time image information such as traffic flow and road conditions. Satellites can provide large-scale, all-weather monitoring capabilities. Sensor networks are distributed in key locations to sense environmental changes and target dynamics in real time.

[0027] It should be noted that in the task effectiveness analysis system involved in the present application, the two parties mentioned can represent the red and blue opposing parties in any task scenario. The task of the present application can be a multi-unmanned aerial vehicle cooperative detection task.

[0028] The decision node is used to process information, analyze battlefield situation, make decision scheme and issue instructions, including control center, joint dispatch center, aircraft with monitoring and dispatching dual functions and information processing center.

[0029] Specifically, the control center and the joint dispatch center are the core places of task command, gathering various types of intelligence information, formulating task plans and instructions through professional analysis and decision-making processes. The aircraft with dual functions of monitoring and dispatching can not only discover abnormal targets in advance, but also make real-time command decisions during flight. The information processing center focuses on the rapid processing and analysis of massive information to provide accurate data support for decision-making.

[0030] The execution nodes are used to perform operational tasks, including automated job equipment (e.g., drones).

[0031] The target nodes are used to identify key locations, including important facility sites, command sites, large buildings, communication hubs, and key production facilities.

[0032] Specifically, important facility sites are important nodes for obtaining key information, and adjusting them can affect the overall information acquisition capability. Command sites are the core of task command, and affecting command sites can cause the task system to be in chaos. Large buildings, communication hubs, and key production facilities are also important supports for task capability, and effective adjustment or protection of them is crucial for achieving task goals.

[0033] S102: Based on the four types of nodes of the OODA loop, determine the actual connection edges, obtain the task network, and extract the generalized OODA loop from the task network to cover the cooperative relationship.

[0034] Based on the four types of nodes of the OODA loop, determine the actual connection edges, including the following.

[0035] When each type of node in the four types of nodes interacts with the four types of nodes including itself, consider both the information sending and receiving directions, and regard the single direction information interaction between each type of node and the four types of nodes including itself as a potential connection edge.

[0036] Determine the actual connection edges based on the potential connection edges.

[0037] Specifically, the present application determines the actual connection edges based on task requirements and logic.

[0038] Extract the generalized OODA loop from the task network to cover the cooperative relationship, including the following.

[0039] The cooperative relationship includes cooperative perception connection edges, cooperative command connection edges, and cooperative execution connection edges.

[0040] The expression of the task network is: . Wherein, is the task network, and the nodes include perception nodes , decision nodes , and execution nodes and target node , actual connection edge including the cooperative perception connection edge from the perception node to the perception node, the decision relationship connection edge from the perception node to the decision node, the first command relationship connection edge from the decision node to the perception node, the cooperative command connection edge from the decision node to the decision node, the second command relationship connection edge from the decision node to the execution node, the cooperative execution connection edge from the execution node to the execution node, the work relationship connection edge from the execution node to the target node and the reconnaissance relationship connection edge from the target node to the perception node. Table 1 is the determined actual connection edge.

[0041] Table 1 Determined actual connection edge

[0042]

[0043] Initialize the counter to 0, which is used to count the target node corresponding to the number of task rings.

[0044] Take the target node as the starting point, and traverse the adjacent nodes through depth-first search, and each time a new node is accessed, the next recursive call is made with the new node as the starting point to continue the depth-first search traversal.

[0045] Record the node set and node type order of the current path.

[0046] In the traversal process, follow the node type constraint.

[0047] The node type constraint includes: the next step from the target node is the perception node . The next step from the perception node is the perception node or the decision node . The next step from the decision node is the decision node or the execution node . The next step from the execution node is the execution node or the target node . Except for the target node , other nodes in the path do not repeat traversal.

[0048] When traversing to the execution node, check whether there is a work relationship connection edge from the execution node to the target node, and if so, add 1 to the value of the counter.

[0049] After each recursive call, the current node is removed from the node set, and other paths are explored, reducing invalid search paths through pruning and backtracking mechanisms to obtain a generalized OODA loop.

[0050] The generalized OODA loop relaxes the restrictions on path structure, allowing more intermediate nodes and branches to be included.

[0051] Specifically, the number of generalized OODA loops starting from the target node is counted in the node set, supporting complex paths that include multiple perception nodes or decision nodes.

[0052] The above process requires traversing all nodes and actual connection edges in the worst case, with a time complexity of , where is the total number of nodes in the task network, is the total number of actual connection edges in the task network. The search space is greatly reduced due to node type constraints and pruning, with efficiency better than full graph depth-first search traversal. The space complexity is , used to store access states and paths.

[0053] S103: Construct a generalized OODA loop node attribute quantification framework based on task requirements, including task-driven node attribute extraction, quantization method, data acquisition path, and data normalization processing.

[0054] Construct a generalized OODA loop node attribute quantification framework based on task requirements, including task-driven node attribute extraction, quantization method, data acquisition path, and data normalization processing, including the following.

[0055] Task-driven node attribute extraction includes: analyzing task intent, analyzing scene requirements, and identifying constraints. Combined with expert experience and data-driven, the core attributes of each node type related to the task are selected. Table 2 shows the core attributes of different types of nodes in different task scenarios.

[0056] Table 2 Core attributes of different types of nodes in different task scenarios

[0057]

[0058] Quantization method and data acquisition path include: adopting five-order quantization criteria, selecting appropriate quantization methods and data acquisition methods according to the type of core attributes: where the five-order quantization criteria include experimental test method, simulation exercise method, historical data statistical analysis, adversarial analysis method and comprehensive integration method.

[0059] Experimental test method is suitable for physical performance indicators, such as detection range, through laboratory or test site data acquisition. Simulation exercise method is suitable for performance indicators, such as operation accuracy, through simulation operation or system confrontation exercise verification. Historical data statistical analysis is suitable for statistical law indicators, such as situation prediction accuracy, mining historical data or expert experience modeling. Confrontation analysis method is suitable for confrontation indicators, such as state stability, through simulation confrontation or environment replication test. Comprehensive integration method is suitable for complex system level indicators, such as cross-region information fusion efficiency, combining multiple methods and data sources analysis. Table 3 is the quantification method and data source of the core attributes of the four types of nodes.

[0060] Table 3 Quantification method and data source of core attributes of four types of nodes

[0061]

[0062] Data normalization processing includes: converting core attribute data of different dimensions or ranges to a unified dimension or range interval.

[0063] Specifically, to ensure the comparability of analysis, some attributes are normalized. The core attributes that need to be normalized are physical values (such as detection range, scaled to [0, 1] range), time-related indicators (such as end-to-end delay, unified dimension), probability indicators (such as information accuracy, converted to a percentage or decimal in a unified dimension), and the method can be selected Min-Max method (range standardization method) or Z-Score method (range standardization method). The core attributes that do not need to be normalized are naturally normalized indicators (such as information accuracy, percentage or score is in the [0, 1] interval).

[0064] S104: Based on the generalized OODA ring node attribute quantification framework, nonlinear aggregation is performed, and the task network performance is comprehensively analyzed based on the probability of task success.

[0065] Based on the generalized OODA ring node attribute quantification framework, nonlinear aggregation is performed, and the task network performance is comprehensively analyzed based on the probability of task success, including the following.

[0066] Based on Obtain the probability of each node type completing the task. Among them, For Or Or Or , The probability of the perception node completing the task is The probability of the decision node completing the task is The probability of the execution node completing the task is The probability of the target node completing the task is The corresponding node is the first a weight of the influencing factor, the first influencing factor of the corresponding node the membership degree or the possibility of the first

[0067] influencing factor of the node to complete the task. Specifically, it is assumed that there are influencing factors affecting the completion of the task by a certain node. For the first influencing factor, the membership degree or the possibility is 1 when it fully meets the task capability requirement, and the membership degree or the possibility is 0 when it does not meet the task capability requirement at all. If the attribute value requirement of the first influencing factor for the specific task is , and the attribute value of the node is , then the membership degree or the possibility of the first influencing factor of the node to complete the task is . This step is to convert into a value between 0 and 1 to represent the degree of satisfaction of the influencing factor to complete the task.

[0068] Based on the probability of each node type to complete the task, the minimum value method is used to aggregate the node set in series to obtain the task completion rate of each node set, and the expression is: . Wherein, is the task completion rate of the first node set, is the number of nodes of the first generalized OODA loop.

[0069] Specifically, for the OODA loop, there can be four nodes, namely the perception node, the decision node, the execution node and the target node. For the generalized OODA loop, the number of nodes is more than 4.

[0070] Based on the task completion rate of each node set, the probability parallel model is used to regard the multiple generalized OODA loops of the target node as a parallel structure to obtain the probability of success in achieving the target, and the expression is: . Wherein, is the probability of success in achieving the first target, is the task completion rate of the first node set, is the total number of node sets that can be formed to achieve the first target.

[0071] Based on the probability of success in achieving the target, the probability of task success is obtained, and the comprehensive analysis of the task network performance is based on the probability of task success, and the expression is: ; wherein, This represents the probability of the task succeeding. The importance of achieving the goal, To achieve the first The probability of achieving a goal The total number required to achieve the goal.

[0072] This application also provides a collaborative task performance analysis device 200 based on ring network fusion, such as... Figure 2 As shown, the device includes: a simplification module 201, an extraction module 202, a construction module 203, and an analysis module 204.

[0073] The simplification module 201 is used to simplify the UAV collaborative mission system into four types of nodes, including an OODA loop. These node types include perception nodes, decision nodes, execution nodes, and target nodes.

[0074] The extraction module 202 is used to determine the actual connection edges based on the four types of nodes of the OODA ring, obtain the task network, extract the generalized OODA ring from the task network, and cover the collaborative relationship.

[0075] Module 203 is used to build a generalized OODA node attribute quantization framework based on task requirements, including task-driven node attribute extraction, quantization methods and data acquisition paths, and data normalization processing.

[0076] Analysis module 204 is used to perform nonlinear aggregation based on the generalized OODA link point attribute quantization framework, and to conduct a comprehensive analysis of the task network performance based on the probability of task success.

[0077] Some modules in the apparatus described in this application can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0078] The apparatus or module described in the above embodiments can be implemented by a computer chip or physical entity, or by a product with a certain function. For ease of description, the above apparatus is described by dividing it into various modules according to their functions. When implementing the embodiments of this application, the functions of each module can be implemented in one or more software and / or hardware. Of course, a module that implements a certain function can also be implemented by combining multiple sub-modules or sub-units.

[0079] The methods, apparatuses or modules described in the present application can be implemented in a computer readable program code manner. The controller can be implemented in any appropriate manner, for example, the controller can take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (for example, software or firmware) executable by the (micro)processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller and an embedded microcontroller. Examples of the controller include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that, in addition to implementing the controller in a pure computer readable program code manner, the same function can also be implemented by logically programming the method steps in the form of logic gates, switches, ASICs, programmable logic controllers and embedded microcontrollers. Therefore, such a controller can be considered as a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can even be considered as both a software module implementing the method and a structure within the hardware component.

[0080] As shown in Figure 3 The embodiment of the present application also provides a collaborative task efficiency analysis server based on ring network fusion, which comprises a memory 301 and a processor 302; the memory 301 is used for storing computer executable instructions; and the processor 302 is used for executing the computer executable instructions to realize the method for analyzing the efficiency of a collaborative task based on ring network fusion.

[0081] The embodiment of the present application also provides a computer readable storage medium, which stores executable instructions, and a computer executes the executable instructions to realize the method for analyzing the efficiency of a collaborative task based on ring network fusion.

[0082] From the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus necessary hardware. Based on such an understanding, the technical solutions of the present application can be embodied in the form of a software product or can be embodied in the implementation process of data migration. The computer software product can be stored in a storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a mobile terminal, a server, or a network device, etc.) to execute the methods described in the embodiments of the present application.

[0083] The various embodiments in the specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments. The whole or part of the present application can be used in a plurality of general or special computer system environments or configurations.

[0084] The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the present application.

Claims

1. A method for analyzing the performance of a collaborative task based on ring network fusion, characterized in that, Comprise: Simplify the unmanned aerial vehicle cooperative task system into four types of node types containing the OODA ring; wherein the node types include perception nodes, decision nodes, execution nodes and target nodes; Based on the four types of nodes of the OODA ring, determine the actual connection edge, obtain the task network, extract the generalized OODA ring from the task network, and cover the cooperative relationship; Build a generalized OODA ring node attribute quantification framework based on task requirements, including task-driven node attribute extraction, quantification method and data acquisition path, and data normalization processing; Based on the generalized OODA ring node attribute quantification framework, nonlinear aggregation is carried out, and the task network efficiency is comprehensively analyzed based on the probability of task success; The generalized OODA ring node attribute quantification framework, nonlinear aggregation is carried out, and the task network efficiency is comprehensively analyzed based on the probability of task success, comprising: based on Obtain the probability of each node type completing the task; where, for or or or , To determine the probability that a sensing node will complete its task. The probability that a decision node will complete its task. The probability that the execution node will complete the task. The probability of completing the task for the target node. For the corresponding node's first The weight of each influencing factor For the corresponding node's first The degree of membership or probability of each influencing factor in completing the task; Based on the probability of each node type completing the task, the minimum value method is used to aggregate the node set in series to obtain the task completion rate of each node set, and the expression is: ; wherein, is the task completion rate of the first node set, is the number of nodes of the first generalized OODA loop. Based on the task completion rate of each node set, the multiple generalized OODA loops of the target node are regarded as a parallel structure by using a probability parallel model, and the probability of achieving the target success is obtained, which is expressed as: ; wherein, is the probability of achieving the success of the first target, is the task completion rate of the first node set, is the total number of node sets that can form the first target. Based on the probability of success in achieving the goal, the probability of success in the task, the comprehensive analysis of the network performance of the task based on the probability of success in the task, the expression is: ; wherein, the probability of success in the task, the importance of achieving the goal, the probability of success in achieving the first goal, the probability of success in achieving the nth goal, the total number of goals achieved.

2. The method of claim 1, wherein the method is based on a ring network fusion collaborative task performance analysis. The perception node is used to obtain environmental information, collect target information and monitor the environment; The decision node is used to process information, analyze the scene situation, formulate a decision scheme and issue an instruction; The execution node is used to execute the operation task; The target node is used to clarify the key position. 3.The method of claim 2, wherein, The four types of nodes of the OODA ring determine the actual connection edge, comprising: When each type of node in the four types of nodes interacts with the four types of nodes including itself, consider the two directions of information sending and receiving, and regard the single direction information interaction between each type of node and the four types of nodes including itself as a potential connection edge; Determine the actual connection edge based on the potential connection edge.

4. The method of claim 3, wherein the method further comprises: The generalized OODA ring is extracted from the task network, covering the cooperative relationship, comprising: The cooperative relationship includes cooperative perception connection edge, cooperative command connection edge and cooperative execution connection edge; The expression of the task network is: ; wherein, is a task network, nodes include a perception node , a decision node , an execution node and a target node , actual connection edges include a cooperative perception connection edge from the perception node to the perception node, a decision relation connection edge from the perception node to the decision node, a first command relation connection edge from the decision node to the perception node, a cooperative command connection edge from the decision node to the decision node, a second command relation connection edge from the decision node to the execution node, a cooperative execution connection edge from the execution node to the execution node, a work relation connection edge from the execution node to the target node and a reconnaissance relation connection edge from the target node to the perception node; initializing a counter to 0 for counting target nodes a corresponding number of task rings; With target node Starting from the node, the system visits adjacent nodes by depth-first search. Each time a new node is visited, the next recursive call is made starting from that new node, and the depth-first search traversal continues. Record the node set and node type order of the current path; In the traversal process, follow the node type constraint; The node type constraint includes: from the target node The next step from the perception node ; The next step from the perception node The next step from the perception node Or the decision node The next step from the decision node The next step from the decision node Or the execution node The next step from the execution node The next step from the execution node Or the target node Except for the target node In the path, other nodes are not repeated. When the execution node is traversed, check whether there is a job relationship connection edge from the execution node to the target node, if there is, add 1 to the value of the counter; After each recursive call, remove the current node from the node set, explore other paths, and through the pruning and backtracking mechanism, reduce invalid search paths to obtain the generalized OODA ring.

5. The method of claim 4, wherein the method further comprises: The generalized OODA ring node attribute quantification framework based on task requirements is constructed, comprising: Task-driven node attribute extraction includes: analyzing the task intention, analyzing the scene demand, and identifying the constraint condition; combined with expert experience and data driven, screen out the core attributes of each node type related to the task; Quantification method and data acquisition path include: adopting five-order quantification criterion, selecting corresponding quantification method and data acquisition method according to the type of core attribute: wherein the five-order quantification criterion includes experimental test method, simulation exercise method, historical data statistical analysis, antagonistic analysis method and comprehensive integration method; Data normalization processing includes: converting the core attribute data of different dimensions or ranges to a unified dimension or range interval.

6. A collaborative task performance analysis device based on ring network fusion, characterized in that, Comprise: Simplify the unmanned aerial vehicle cooperative task system into four types of node types containing the OODA ring; wherein the node types include perception nodes, decision nodes, execution nodes and target nodes; The extraction module is configured to determine actual connection edges based on four types of nodes of the OODA loop, obtain a task network, extract a generalized OODA loop from the task network, and cover a cooperative relationship. The construction module is configured to construct a node attribute quantification framework of the generalized OODA loop based on a task demand, including task-driven node attribute extraction, a quantification method, a data acquisition path, and data normalization processing. The analysis module is configured to perform nonlinear aggregation based on the node attribute quantification framework of the generalized OODA loop, and comprehensively analyze the task network performance based on a probability of task success. The analysis module is configured to perform nonlinear aggregation based on the node attribute quantification framework of the generalized OODA loop, and comprehensively analyze the task network performance based on a probability of task success. based on Obtain the probability of each node type completing the task; where, for or or or , To determine the probability that a sensing node will complete its task. The probability that a decision node will complete its task. The probability that the execution node will complete the task. The probability of completing the task for the target node. For the corresponding node's first The weight of each influencing factor For the corresponding node's first The degree of membership or probability of each influencing factor in completing the task; Based on the probability of each node type completing the task, the minimum value method is used to aggregate the node set in series to obtain the task completion rate of each node set, and the expression is: ; wherein, is the task completion rate of the th node set, is the number of nodes of the th generalized OODA loop; Based on the task completion rate of each node set, the multiple generalized OODA loops of the target node are regarded as a parallel structure by using a probability parallel model, and the probability of achieving the target success is obtained, which is expressed as: ; wherein, is the probability of achieving the first target success, is the task completion rate of the first node set, is the total number of node sets that can be formed to achieve the first target. Based on the probability of success in achieving the goal, the probability of success in the task, the comprehensive analysis of the network performance of the task based on the probability of success in the task, the expression is: ; wherein, the probability of success in the task, the importance of achieving the goal, the probability of success in achieving the first goal, the total number of goals achieved.

7. A collaborative task performance analysis server based on ring network fusion, characterized in that, The computer readable storage medium stores executable instructions, and the computer executes the executable instructions to implement the method of any one of claims 1-5. The computer readable storage medium stores executable instructions, and the computer executes the executable instructions to implement the method of any one of claims 1-5. ​ 8. A computer-readable storage medium, characterized in that, ​

Images