A simulation combat resource management system and method based on multi-agent system

Abstract

A simulated combat resource management system and method based on a multi-agent system, comprising a multi-agent system layer, an environment simulation layer and a management decision layer; a reconnaissance agent collects multi-source battlefield environment data through multiple sensors, and a multi-source data fusion module generates battlefield fusion data from the multi-source battlefield environment data acquired by the reconnaissance agent; a battlefield situation analysis module performs battlefield situation analysis based on the battlefield fusion data, a decision suggestion module is used to generate a multi-level decision tree based on the battlefield situation analysis result and send it to the command agent to generate a simulated combat resource management plan, and the firepower agent and the support agent execute the simulated combat resource management plan. The present invention realizes real-time dynamic management of combat resources and improves resource utilization efficiency through the collaboration and autonomy of the multi-agent system, multi-source data fusion and multi-level decision trees.

Description

Technical Field

[0001] The present invention relates to the field of resource management technology, and more specifically, to a simulation combat resource management system and method based on a multi-agent system. Background Art

[0002] Combat resource management in modern warfare is a complex and dynamic process. Traditional static management methods can no longer meet the needs of modern warfare. With the increasing complexity and uncertainty of the battlefield environment, modern warfare has higher and higher requirements for resource management. Traditional resource management methods rely on manual decision-making and centralized control, which are difficult to cope with the dynamically changing battlefield situation.

[0003] Prior art 1: The centralized database system is used to manage combat resources through centralized data storage and processing. The centralized system has a simple structure, centralized data storage, and is easy to manage and maintain. However, this method mainly relies on manual input and data update, and has problems such as slow response speed and poor data real-time performance. Especially under high-intensity and high-frequency data update requirements, the processing capacity of the centralized system is obviously insufficient, resulting in poor data real-time performance and delayed decision-making. It shows obvious performance bottlenecks when processing large amounts of real-time data, and cannot meet the high-frequency data update requirements of modern warfare.

[0004] The main implementation methods of the centralized resource management system include data acquisition module, data storage module and data analysis module. The data acquisition module obtains battlefield information through sensors, the data storage module centrally manages this information, and the data analysis module performs information processing and decision support. The existing centralized resource management system obtains battlefield information through the data acquisition module and stores this information in a database. The data analysis module processes the stored data and generates decision support information. However, this method has obvious bottlenecks in data update and processing speed, and cannot meet the real-time and high-efficiency requirements of modern operations.

[0005] Existing Technology 2: Some advanced combat resource management systems have begun to introduce distributed databases and preliminary intelligent analysis technologies. Distributed database technology improves the scalability and fault tolerance of the system through decentralized storage and processing of data, but it is still insufficient in terms of intelligent collaboration and real-time dynamic management.

[0006] The implementation scheme of distributed database technology includes distributed data storage, data synchronization mechanism and basic intelligent analysis modules. The distributed management and preliminary intelligent processing of data are realized through the collaborative work of these modules. The distributed database system realizes the decentralized management and processing of data by storing data on multiple nodes. This method improves the scalability and fault tolerance of the system to a certain extent and reduces the risk of single point failure. However, in terms of intelligent collaboration and real-time dynamic management, the existing technology still has shortcomings and is difficult to meet the needs in complex battlefield environments.

[0007] Therefore, the problems existing in the prior art need to be further improved and developed. Summary of the invention

[0008] (I) Purpose of the invention: In order to solve the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide a simulated combat resource management system and method based on a multi-agent system.

[0009] (II) Technical solution: In order to solve the above technical problems, the present technical solution provides a simulated combat resource management system based on a multi-agent system, comprising a multi-agent system layer, an environmental simulation layer and a management decision layer; the multi-agent system layer comprises a command agent, a communication agent, a reconnaissance agent, a fire agent and a support agent interconnected through a communication network; the environmental simulation layer comprises a multi-source data fusion module, a real-time data update module and an environmental database; the management decision layer comprises a battlefield situation analysis module and a decision suggestion module;

[0010] The reconnaissance agent collects multi-source battlefield environment data through multiple sensors, and the multi-source data fusion module uses weighted average fusion to fuse the multi-source battlefield environment data acquired by the reconnaissance agent by assigning weights to battlefield environment data from different data sources, and generates battlefield fusion data by performing multi-source data fusion;

[0011] The battlefield situation analysis module performs battlefield situation analysis based on battlefield fusion data. The decision suggestion module is used to generate a multi-level decision tree based on the battlefield situation analysis results and send it to the command intelligent agent to generate a simulated combat resource management plan. The firepower intelligent agent and the support intelligent agent execute the simulated combat resource management plan.

[0012] A simulated combat resource management system based on a multi-agent system, wherein the command agent, the communication agent, the reconnaissance agent, the firepower agent and the support agent are mutually independent subsystems, adopt a distributed network architecture and are distributed on different combat entities;

[0013] The command agent includes a command and coordination module and a plan generation module. The command and coordination module commands and coordinates the actions of other agents, and the plan generation module generates a simulated combat resource management plan.

[0014] The communication agent includes a data communication module and a network management module. The data communication module establishes and manages data transmission links with other agents, and the network management module supports self-repair and self-optimization of the network.

[0015] The reconnaissance agent includes a sensor module, a target recognition module and a data processing module. The sensor module includes a sensor network composed of multiple sensors for sensing the battlefield environment and collecting battlefield environment data. The target recognition module identifies enemy targets and combat resources. The data processing module pre-processes the battlefield environment data collected by the sensor module.

[0016] The firepower agent includes a firepower resource database and a firepower scheduling module. The firepower resource data stores and manages information of various firepower resources. The firepower scheduling module schedules firepower resources to perform strike missions.

[0017] The support intelligent body includes a logistics resource database and a logistics scheduling module. The logistics resource database stores and manages information of various logistics resources. The logistics scheduling module schedules logistics resources to support combat operations.

[0018] A simulated combat resource management system based on a multi-agent system, wherein the process of the multi-source data fusion module fusing the multi-source battlefield environment data acquired by the reconnaissance agent comprises:

[0019] Divide the multi-source battlefield environment data collected by multiple sensors received by the sensor module of the reconnaissance agent into M data sources, where M is the number of different battlefield environment data types received by the sensor module, where M≥2;

[0020] Divide the battlefield into N battlefield areas, namely, the first battlefield area, the second battlefield area, the third battlefield area, ..., the Nth battlefield area, where N ≥ 2;

[0021] The battlefield environment data from multiple different data sources are numbered and then integrated into multi-source battlefield environment data D1, D2, D3, ..., D M , where D1, D2, D3, …, D M Represents battlefield environment data from different data sources respectively; battlefield environment data D of the Mth data source in multi-source battlefield environment data M =[D M1 , D M2 , D M3 , …, D MN], where D M1 , D M2 , D M3 , …, D MN They are battlefield environment data of different battlefield areas in the Mth data source;

[0022] Weights are assigned to multi-source battlefield environment data, and different weight values ​​W are assigned to different data sources, where the weight value W=[W1, W2, W3, ..., W M ], where W1, W2, W3, …, W M Represent the weights of data from different data sources, W1+W2+W3+…+W M =1;

[0023] The multi-source battlefield data from different data sources are fused by weighted average fusion technology to obtain fused data D 融合 , the fused data D 融合 =[W1*D1,W2*D2,W3*D3…W M *D M ], where the data from different data sources and their weights correspond one to one.

[0024] A simulated combat resource management system based on a multi-agent system, wherein the multi-level decision tree includes an upper-level firepower resource decision tree and a lower-level logistics resource decision tree, the firepower resource decision tree includes three decision nodes: threat assessment, firepower resource sufficiency, and reinforcement firepower completion time; the logistics resource decision tree includes three decision nodes: logistics resource sufficiency, front-line demand, and path optimization.

[0025] A simulated combat resource management system based on a multi-agent system, in which a firepower resource decision tree performs threat assessment through battlefield fusion data and calculates the threat index of enemy units to our targets. The calculation formula of the threat index is: T=α×d+β×v, where T is the threat index, the d parameter is the distance between the enemy unit and our target collected by the sensor module, the v parameter is the enemy's moving speed collected by the sensor module, and α and β are parameter weights.

[0026] A simulated combat resource management system based on a multi-agent system, wherein the resource management plan includes a firepower resource management plan and a logistics resource scheduling plan; the firepower resource management plan includes a target area sub-plan, a weapon type sub-plan, and a strike intensity and frequency sub-plan, and the logistics resource scheduling plan includes a priority supply area sub-plan, a material type and quantity sub-plan, and a transportation path sub-plan.

[0027] A simulated combat resource management system based on a multi-agent system, which also includes an optimization and adjustment layer, which includes a solution evaluation module and a real-time monitoring module; the real-time monitoring module monitors the battlefield environment in real time based on battlefield environment data in an environment database; the solution evaluation module evaluates the applicability of the current resource management solution based on the real-time battlefield environment data according to a suitability evaluation index system, and determines whether the resource management solution needs to be optimized and adjusted.

[0028] A simulated combat resource management system based on a multi-agent system, wherein the applicability evaluation index system includes three levels: an overall goal, a primary index and a secondary index; the overall goal of the applicability evaluation index system is a suitability score, which reflects the overall effectiveness of a resource management plan; the primary index includes five primary index items: a task achievement index, a resource utilization index, a battlefield control index, an adaptability and flexibility index, and a sustainability index; the task achievement index includes three secondary index items: a firepower strike accuracy index, a logistics supply punctuality index, and an enemy containment rate index; the resource utilization index includes three secondary index items: a firepower resource utilization index, a logistics resource utilization index, and a resource scheduling error rate index; the battlefield control index includes three secondary index items: an enemy containment rate index, a key zone control rate index, and an enemy advancement suppression rate index; the adaptability and flexibility index includes three secondary index items: a material demand adaptability index, an enemy dynamic adaptability index, and a real-time feedback adjustment capability index; and the adaptability and flexibility index includes two secondary index items: a battlefield persistence index and a resource continuous supply capability index.

[0029] A simulated combat resource management system based on a multi-agent system, wherein the secondary indicator score is calculated by actual execution effect and plan achievement rate;

[0030] The calculation formula for the applicability score of the first-level indicator is: , where m is the total number of secondary index items, 1≤n≤m, is the secondary indicator score for the corresponding secondary indicator item, is the weight of each secondary indicator item;

[0031] The calculation formula for the overall target suitability score is: ,

[0032] Where j is the total number of first-level index items, 1≤k≤j, is the first-level indicator score for the corresponding first-level indicator item, is the weight of each first-level indicator in the overall applicability.

[0033] The present invention also provides a method for managing simulated combat resources based on a multi-agent system, which is used to implement any of the above-mentioned simulated combat resource management systems based on a multi-agent system, comprising the following steps:

[0034] Step 1: The sensor module of the reconnaissance agent collects multi-source battlefield environment data collected by multiple sensors, and transmits the data to the environment simulation layer through the data communication module of the communication agent unit. The multi-source data fusion module of the environment simulation layer fuses the multi-source battlefield environment data from different sensors and saves them in the environment database;

[0035] Step 2: The battlefield situation analysis module of the management decision-making layer performs battlefield situation analysis based on the battlefield environment data in the environment database, the firepower resource data in the firepower resource database, and the logistics resource data in the logistics database. The decision suggestion module generates a multi-level decision tree based on the battlefield situation analysis. The solution generation module of the command agent generates a preliminary resource management solution based on the battlefield situation analysis results and the multi-level decision tree.

[0036] Step 3: The command and coordination module of the command agent generates corresponding instructions according to the resource management plan and sends the instructions to the firepower scheduling module of the firepower agent and the logistics scheduling module of the support agent to execute the resource management plan and perform combat resource management;

[0037] Step 4. The real-time data update module inputs the real-time battlefield environment data into the environment database. The real-time monitoring module of the optimization and adjustment layer monitors the battlefield environment data in real time. The program evaluation module evaluates the applicability of the current resource management program based on the applicability evaluation index system with the real-time battlefield environment data. When the applicability score is less than 80 points, the battlefield situation analysis module and the real-time monitoring module of the management decision-making layer are reactivated to regenerate the optimized multi-level decision tree. The command agent regenerates the resource management plan based on the optimized multi-level decision tree, and sends it to the firepower scheduling module of the firepower agent and the logistics scheduling module of the support agent for plan execution.

[0038] (III) Beneficial effects: The present invention provides a simulated combat resource management system and method based on a multi-agent system. Through the collaboration and autonomy of the multi-agent system, real-time dynamic management of combat resources is achieved to improve resource utilization efficiency. Through multi-source data fusion, accurate decision support is provided, resource allocation is optimized, and combat effectiveness is improved. Through a multi-level decision tree and a suitability evaluation index system, the system can provide accurate decision support information, optimize resource allocation, and improve combat effectiveness in a complex and changeable battlefield environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the overall architecture of a simulated combat resource management system based on a multi-agent system of the present invention;

[0040] Figure 2 It is a schematic diagram of the composition structure of a multi-agent system of a simulated combat resource management system based on a multi-agent system of the present invention;

[0041] Figure 3 It is a schematic diagram of a process of data fusion performed by a multi-source data fusion module of a simulated combat resource management system based on a multi-agent system of the present invention;

[0042] Figure 4 It is a decision process diagram of a multi-level decision tree of a simulated combat resource management method based on a multi-agent system of the present invention;

[0043] Figure 5 It is a schematic diagram of the steps of a method for simulating combat resource management based on a multi-agent system of the present invention. DETAILED DESCRIPTION

[0044] The present invention is further described in detail below in conjunction with preferred embodiments. More details are elaborated in the following description to facilitate a full understanding of the present invention. However, the present invention can obviously be implemented in a variety of other ways different from the description. Those skilled in the art can make similar generalizations and deductions based on actual application situations without violating the connotation of the present invention. Therefore, the protection scope of the present invention should not be limited by the content of this specific embodiment.

[0045] The accompanying drawings are schematic diagrams of embodiments of the present invention. It should be noted that the drawings are only examples and are not drawn to scale, and should not be used to limit the actual protection scope of the present invention.

[0046] The present invention provides a simulation combat resource management system based on a multi-agent system. Figure 1 As shown, it includes a multi-agent system layer, an environment simulation layer, a management decision layer, and an optimization and adjustment layer.

[0047] Preferably, the multi-agent system layer includes a command agent, a communication agent, a reconnaissance agent, a firepower agent, and a support agent interconnected through a communication network. Each of the agents is an independent subsystem that can perform specific tasks in its independent operating environment. Each of the agents adopts a distributed network architecture, is distributed on different combat entities, and is interconnected through a communication network to form a whole.

[0048] The present invention is preferably, Figure 2As shown, the command agent includes a command and coordination module and a plan generation module. The command and coordination module commands and coordinates the actions of other agents to ensure coordinated operations between them; the plan generation module generates a simulated combat resource management plan based on the current battlefield situation and mission requirements.

[0049] The communication agent includes a data communication module and a network management module. The data communication module is responsible for establishing and managing data transmission links with other agents, realizing data encapsulation, transmission, routing and decapsulation, and ensuring accurate data transmission between different agents; the network management module manages and maintains the communication network, supports network self-repair and self-optimization, and ensures network connectivity and communication reliability.

[0050] The reconnaissance agent collects multi-source battlefield environment data through multiple sensors, including sensor modules, target recognition modules and data processing modules. The sensor module includes a sensor network composed of multiple sensors such as radar, infrared detectors, cameras, etc., which is used to perceive the battlefield environment and collect battlefield environment data. The target recognition module analyzes and processes the battlefield environment data collected by the sensor module to identify enemy targets and combat resources; the data processing module pre-processes, converts the format and normalizes the battlefield environment data collected by the sensor module, extracts useful intelligence information, and passes this information to the environment simulation layer.

[0051] The firepower intelligent agent includes a firepower resource database and a firepower scheduling module. The firepower resource database is used to store and manage information on various firepower resources, including weapon type, quantity, location, performance parameters, etc.; the firepower scheduling module schedules firepower resources to perform strike tasks according to the instructions of the command intelligent agent and the current battlefield situation, including selecting weapon types, determining strike targets, planning firepower strike sequence and intensity, etc.

[0052] The support agent includes a logistics resource database and a logistics dispatch module. The logistics resource database is used to store and manage information on various logistics resources, including ammunition, fuel, food, medical supplies, etc. The logistics dispatch module dispatches logistics resources to support combat operations according to the instructions of the command agent and the current battlefield situation, including functions such as material transportation, distribution, repair and maintenance.

[0053] Preferably, the environmental simulation layer includes a multi-source data fusion module, a real-time data update module and an environmental database. The multi-source data fusion module performs multi-source data fusion on the battlefield environmental data acquired by the reconnaissance agent, integrates the battlefield environmental data provided by the reconnaissance agent, and improves the accuracy and integrity of the battlefield environmental data; the environmental database stores and manages the battlefield environmental data, and provides basic data support for the system; the real-time data update module updates the battlefield environmental data in real time according to the sensor module of the reconnaissance agent, and inputs the real-time battlefield environmental data into the environmental database, ensuring the real-time nature of the battlefield environmental data and providing support for decision-making.

[0054] Preferably, the management decision layer includes a battlefield situation analysis module and a decision suggestion module. The battlefield situation analysis module performs battlefield situation analysis based on the environment database and sends the battlefield situation analysis result to the decision suggestion module, the decision suggestion module generates a multi-level decision tree based on the battlefield situation analysis result, and the solution generation module of the command intelligent agent generates a resource management solution based on the multi-level decision tree.

[0055] Preferably, the optimization and adjustment layer includes a solution evaluation module and a real-time monitoring module. The real-time monitoring module monitors the battlefield environment in real time according to the battlefield environment data in the environment database; the solution evaluation module evaluates the applicability of the current resource management solution according to the applicability evaluation index system based on the real-time battlefield environment data, and determines whether the resource management solution needs to be optimized and adjusted.

[0056] The environmental simulation layer includes a multi-source data fusion module for fusing the multi-source battlefield environment data acquired by the reconnaissance agent, a real-time data update module for ensuring the real-time nature of the battlefield environment data, and an environmental database for storing and managing the battlefield environment data; the management decision-making layer includes a battlefield situation analysis module for performing battlefield situation analysis based on the environmental database and a decision suggestion module for generating a multi-level decision tree based on the battlefield situation analysis results and sending it to the command agent to generate a simulated resource management plan; the optimization and adjustment layer includes a plan evaluation module and a real-time monitoring module; the real-time monitoring module monitors the battlefield environment in real time based on the battlefield environment data in the environmental database; the plan evaluation module evaluates the applicability of the current resource management plan based on the real-time battlefield environment data according to the applicability evaluation index system, and determines whether the resource management plan needs to be optimized and adjusted.

[0057] Preferably, the reconnaissance agent collects multi-source battlefield environment data through multiple sensors, and the multi-source data fusion module uses weighted average fusion to fuse the multi-source battlefield environment data acquired by the reconnaissance agent by assigning weights to battlefield environment data from different data sources, and generates battlefield fusion data by performing multi-source data fusion;

[0058] The battlefield situation analysis module performs battlefield situation analysis based on battlefield fusion data. The decision suggestion module is used to generate a multi-level decision tree based on the battlefield situation analysis results and send it to the command intelligent agent to generate a simulated combat resource management plan. The firepower intelligent agent and the support intelligent agent execute the simulated combat resource management plan.

[0059] The present invention is preferably, Figure 3 As shown, the multi-source data fusion module performs multi-source data fusion on the multi-source battlefield environment data acquired by the reconnaissance agent, including:

[0060] S101: Divide the multi-source battlefield environment data received by the sensor module of the reconnaissance agent into M data sources, where M is the number of different battlefield environment data types received by the sensor module, where M≥2.

[0061] S102: The reconnaissance agent divides the battlefield into N battlefield areas, namely, a first battlefield area, a second battlefield area, a third battlefield area, ..., an Nth battlefield area, where N≥2.

[0062] S103: The battlefield environment data from multiple different data sources are numbered and then integrated into multi-source battlefield environment data D1, D2, D3, ..., D M , where D1, D2, D3, …, D M Represents battlefield environment data from different data sources respectively; battlefield environment data D of the Mth data source in multi-source battlefield environment data M =[D M1 , D M2 , D M3 , …, D MN ], where D M1 , D M2 , D M3 , …, D MN They are respectively the battlefield environment data of different battlefield areas in the Mth data source.

[0063] S104: Weighting the multi-source battlefield environment data, and assigning different weight values ​​W to different data sources, where the weight value W=[W1, W2, W3, ..., W M ], where W1, W2, W3, …, W M Represent the weights of data from different data sources, W1+W2+W3+…+W M =1.

[0064] S105: Use weighted average fusion technology to fuse multi-source battlefield data from different data sources to obtain fused data D 融合 , the fused data D 融合 =[W1*D1,W2*D2,W3*D3…WM *D M ], where the data from different data sources and their weights correspond one to one.

[0065] In a first preferred embodiment of the present invention, the battlefield is divided into five battlefield areas, and the sensor module receives multi-source data from two sources, namely enemy density data D1 detected by radar and enemy movement speed data D2 monitored by drones.

[0066] The enemy density data detected by radar in the first battlefield area is 0.9, the enemy density data detected by radar in the second battlefield area is 0.8, the enemy density data detected by radar in the third battlefield area is 0.6, the enemy density data detected by radar in the fourth battlefield area is 0.4, and the enemy density data detected by radar in the fifth battlefield area is 0.7. The weight of the enemy density data detected by radar is 70%.

[0067] The enemy's movement speed data in the first battlefield area monitored by the drone is 0.5, the enemy's movement speed data in the first battlefield area monitored by the drone is 0.6, the enemy's movement speed data in the first battlefield area monitored by the drone is 0.4, the enemy's movement speed data in the first battlefield area monitored by the drone is 0.5, and the enemy's movement speed data in the first battlefield area monitored by the drone is 0.6. The weight of the enemy's movement speed data monitored by the drone is 30%.

[0068] That is, the number of battlefield areas N=5, the number of data sources M=2, the enemy density data detected by radar D1 = [0.9, 0.8, 0.6, 0.4, 0.7], and the enemy movement speed data monitored by drone D2 = [0.5, 0.6, 0.4, 0.3, 0.6]; the weight of the enemy density data D1 detected by radar is W1=0.7, and the weight of the enemy movement speed data D2 monitored by drone is W2=0.3.

[0069] The enemy density data D1 detected by radar and the enemy movement speed data D2 monitored by drone are fused by weighted average fusion technology to obtain fused data D 融合 .

[0070] D 融合 =D1*0.7+D2*0.3= [0.78, 0.74, 0.54, 0.37, 0.67].

[0071] The present invention is preferably a second preferred embodiment of the present invention, the decision suggestion module generates a multi-level decision tree according to the battlefield situation analysis result, such as Figure 4As shown, the multi-level decision tree includes an upper-level firepower resource decision tree and a lower-level logistics resource decision tree. The firepower resource decision tree includes three decision nodes: threat assessment, firepower resource sufficiency, and reinforcement firepower completion time; the logistics resource decision tree includes three decision nodes: logistics resource sufficiency, front-line demand, and path optimization.

[0072] Specifically, first, the battlefield situation analysis module obtains the fusion data of the battlefield environment data and performs threat assessment. The threshold of the threat index is set to 15. If the threat index is less than or equal to 15, the battlefield area is determined to be a low-threat area, and a decision suggestion of "maintaining the existing firepower configuration" is generated; if the threat index is greater than 15, a decision suggestion of "allocating firepower resources to high-threat areas" is generated. Next, the firepower resources are determined to be sufficient according to the data in the acquired firepower resource database. If the firepower resources are sufficient, the decision suggestion of "allocating firepower resources to high-threat areas" is executed, and a decision suggestion of "marking the area as the main strike target" is generated; if the firepower resources are insufficient, a decision suggestion of "applying for reinforcements" is generated. Next, the system determines whether the reinforcement firepower can be completed within the specified time according to the current data in the firepower resource database. If the reinforcement firepower can be achieved, a decision suggestion of "allocating firepower resources from the rear" is generated; if the reinforcement firepower cannot be achieved, a decision suggestion of "unable to allocate firepower, maintain defense and dispatch logistics resources" is generated. The above is the first-level decision tree process. After completing the first-level decision tree process, the second-level decision tree process is entered.

[0073] Specifically, the battlefield situation analysis module determines whether the logistics resources are sufficient based on the logistics resource data in the logistics database. If the logistics resources are sufficient, the battlefield fusion data is then used to determine whether the frontline demand is high. If the frontline demand is high, a decision suggestion of "giving priority to allocating logistics resources to the frontline" is generated. If the frontline demand is not high, a decision suggestion of "maintaining the established allocation plan of logistics resources" is generated. If the logistics resources are insufficient, a decision suggestion of "replanning the logistics supply route and optimizing transportation resources" is generated. Next, it is determined whether the path optimization is successful. If the path optimization is successful, a decision suggestion of "ensuring the rapid transportation of logistics resources to the demand area" is generated. If the path optimization is unsuccessful, a decision suggestion of "calling logistics resources from other areas as an emergency" is generated.

[0074] Afterwards, the plan generation module of the command agent generates a preliminary resource management plan based on the multi-level decision tree, and sends it to the firepower agent and the support agent to execute it.

[0075] Preferably, the threat index is calculated as follows: T=α×d+β×v, where T is the threat index, the d parameter is the distance between the enemy unit and our target (such as defense line, firepower resources, etc.) collected by the sensor module, the v parameter is the enemy's moving speed collected by the sensor module, and α and β are parameter weights that can be adjusted according to current combat requirements. The threat index is used to help determine whether it is necessary to concentrate firepower or increase logistical support.

[0076] The present invention preferably uses a genetic algorithm to optimize the route, reduce transportation time and cost, and ensure that supplies can quickly reach the front-line troops.

[0077] The process of using genetic algorithms to perform path optimization includes:

[0078] 1. Initial population generation: Randomly generate a set of possible transportation path sequences, such as A->B->C->D->E.

[0079] 2. Fitness evaluation: Calculate the total distance of each path as the fitness value. The shorter the path, the higher the fitness.

[0080] 3. Selection operation: select individuals for reproduction based on fitness values, giving priority to paths with higher fitness.

[0081] 4. Crossover operation: Generate new candidate paths by partially exchanging sub-paths in the parent path.

[0082] 5. Mutation operation: Randomly change the order of sites in the path to increase population diversity and prevent local optimal solutions.

[0083] 6. Iterative optimization: After multiple iterations, the optimal path is found to ensure that transportation costs and time are minimized.

[0084] Preferably, in the present invention, the resource management plan in step 3 includes a firepower resource management plan, a logistics resource scheduling plan, a command and coordination plan, etc.

[0085] In a third preferred embodiment of the present invention, as shown in Table 1, the resource management scheme includes a firepower resource management scheme, a logistics resource scheduling scheme, and a command and coordination scheme. The firepower resource management scheme includes a target area sub-scheme, a weapon type sub-scheme, and a strike intensity and frequency sub-scheme; the logistics resource scheduling scheme includes a priority supply area sub-scheme, a material type and quantity sub-scheme, and a transportation path sub-scheme; and the command and coordination scheme includes a coordination instruction sub-scheme, an execution time sub-scheme, and a feedback mechanism sub-scheme. The target area sub-scheme is specifically the coordinates of the area where concentrated firepower is needed; the weapon type sub-scheme is specifically the type of weapon that needs to be selected, such as artillery, missiles, drones, etc.; the strike intensity and frequency sub-scheme is specifically the intensity of the firepower unit's strike and the time interval for executing firepower strikes; the priority supply area sub-scheme is specifically the area where material supply needs to be given priority, such as the area where material supply is urgently needed on the front line; the material type and quantity sub-scheme is specifically the type and quantity of the required supply materials, such as ammunition, food, medical supplies, etc.; the transportation path sub-scheme is specifically the optimal transportation path to ensure the rapid arrival of materials; the coordination instruction sub-scheme is specifically the specific execution instructions issued to each firepower intelligent agent and / or support intelligent agent; the execution time sub-scheme is specifically the execution time of each resource management instruction; the feedback mechanism sub-scheme is specifically a real-time feedback mechanism to record the execution status and battlefield changes to ensure timely adjustment of the resource management plan.

[0086] Table 1 Specific examples of resource management solutions

[0087]

[0088] Preferably, in the present invention, the applicability evaluation index system in the optimization and adjustment layer includes three levels: overall goal, primary index and secondary index. The overall goal of the applicability evaluation index system is the applicability score (0-100 points), which reflects the overall effectiveness of the resource management plan; the first-level index includes five first-level index items: task achievement index, resource utilization index, battlefield control index, adaptability and flexibility index, and sustainability index; the task achievement index further includes three second-level index items: firepower strike accuracy index, logistics supply punctuality index, and enemy containment rate index; the resource utilization index further includes three second-level index items: firepower resource utilization index, logistics resource utilization index, and resource scheduling error rate index; the battlefield control index further includes three second-level index items: enemy containment rate index, key zone control rate index, and enemy advancement suppression rate index; the adaptability and flexibility index further includes three second-level index items: material demand adaptability index, enemy dynamic adaptability index, and real-time feedback adjustment capability index; the adaptability and flexibility index further includes two second-level index items: battlefield persistence index and resource continuous supply capability index, so as to more accurately evaluate the various execution effects of the plan. The indicator description of each second-level indicator item and the calculation formula of the second-level indicator score are shown in Table 2.

[0089] Table 2: Calculation formula for the secondary indicator score of the applicability evaluation index system

[0090]

[0091] The secondary indicator score of the present invention is calculated by the actual execution effect and the plan achievement rate. A weight is set for each secondary indicator, and a weighted scoring algorithm is used to generate a primary indicator applicability score. A weight is set for each primary indicator, and a weighted scoring algorithm is used to generate an overall goal applicability score. If the score is less than 80 points, the system will determine that the current solution is not applicable and needs to be optimized.

[0092] The calculation formula for the applicability score of the first-level indicator is: , where m is the total number of secondary index items, 1≤n≤m, is the secondary indicator score for the corresponding secondary indicator item, is the weight of each secondary indicator item.

[0093] For example, if the firepower strike accuracy index in the secondary index is 80 points and the weight is 0.5, the logistics supply punctuality index is 70 points and the weight is 0.3, and the key task completion rate index is 90 points and the weight is 0.2, then the corresponding first-level indicator's task achievement index's applicability score = (80×0.5)+(70×0.3)+(90×0.2)=79.

[0094] The calculation formula for the overall target suitability score is: ,

[0095] Where j is the total number of first-level index items, 1≤k≤j, is the first-level indicator score for the corresponding first-level indicator item, is the weight of each first-level indicator in the overall applicability.

[0096] If the overall target suitability score is less than 80 points, the battlefield situation analysis module and real-time monitoring module of the management decision-making layer will be reactivated to regenerate the optimized multi-level decision tree. The command agent will regenerate the resource management plan according to the optimized multi-level decision tree and send it to each agent for execution.

[0097] Preferably, the optimization and adjustment layer also includes a data breakpoint module, which monitors the values ​​of the storage addresses of the multi-source battlefield environment data collected by each sensor in the environment simulation layer during system runtime. When these values ​​change unexpectedly, the system will suspend execution at the address and check the call stack and other context information at that time to determine the cause of the address value change (such as sensor damage, communication interference, power supply problem, or intrusion tampering, etc.), and send the cause to the command intelligent agent, which will formulate a recovery plan and release the suspension after the recovery plan is completed.

[0098] The data breakpoint module includes an address value monitoring component, a breakpoint setting component, a change detection component, a breakpoint analysis component, and a breakpoint recovery component.

[0099] The address value monitoring component is responsible for real-time monitoring of the memory address values ​​of sensor data, ensuring that the system can continuously track the data stored in these addresses and quickly respond to any changes. The reconnaissance agent collects multi-source battlefield environment data through sensors. These data are stored in the environment database of the environment simulation layer. The address value monitoring component monitors the memory address values ​​of these data to track data changes in real time. The real-time data update module ensures that the data in the environment database is up to date, providing an accurate data monitoring basis for the data breakpoint module.

[0100] The breakpoint setting component sets breakpoints at key sensor data addresses, and when the value of the sensor data changes, these breakpoints can trigger execution suspension for further analysis.

[0101] The change detection component is used to identify and verify unexpected changes in sensor data, including sudden jumps, continuous deviations or complete loss of detection data, and compares it with the detection threshold preset by the change detection component to confirm whether it is an unexpected change. The system pauses execution at the address where the unexpected data change occurs to perform detailed diagnosis.

[0102] The breakpoint analysis component is responsible for collecting and analyzing the current call stack, system logs, sensor status and other contextual information to determine the possible causes of the change (sensor damage, communication interference, power supply problems or intrusion tampering, etc.) and the actual address of the sensor corresponding to the address where the breakpoint occurs. Once an unexpected change is detected, the breakpoint analysis component will be activated.

[0103] The breakpoint recovery component sends the analysis results and inferred causes of the breakpoint analysis component to the command intelligent agent so that it can formulate a corresponding recovery plan according to the situation; the command intelligent agent formulates and implements the best breakpoint recovery plan based on the received information, such as mobilizing the firepower intelligent agent to provide battlefield cover, mobilizing the support intelligent agent to replace the backup sensor for the sensor with the actual address corresponding to the address where the breakpoint occurred, or mobilizing the support intelligent agent to perform line repair on the line where the network failure occurred; once the recovery plan is completed and verified to be effective, the breakpoint recovery component will release the suspended execution operation and restore the normal operation of the system.

[0104] The data breakpoint module of the present invention shares the monitoring results of sensor data with the solution evaluation module. When the data breakpoint module detects unexpected changes, it notifies the solution evaluation module, and then the solution evaluation module re-evaluates the applicability of the current resource management solution and determines whether the resource management solution needs to be optimized and adjusted based on the information provided by the data breakpoint module to adapt to changes in the battlefield environment.

[0105] The analysis results of the data breakpoint module are also fed back to the real-time monitoring module to update the monitoring strategy and improve the monitoring sensitivity, especially when potential sensor damage or communication interference is detected.

[0106] Through this design, the data breakpoint module provides a powerful fault detection and recovery mechanism for the simulated combat resource management system, ensuring that the system can remain stable and reliable in the face of various challenges. This mechanism not only improves the robustness of the system, but also enhances its adaptability and sustainability in complex battlefield environments.

[0107] Based on the above-mentioned simulated combat resource management system based on a multi-agent system, the present invention also provides a simulated combat resource management method based on a multi-agent system, such as Figure 5 As shown, the following steps are included:

[0108] Step 1: The sensor module of the reconnaissance agent collects multi-source battlefield environment data collected by multiple sensors, and transmits the data to the environment simulation layer through the data communication module of the communication agent unit. The multi-source data fusion module of the environment simulation layer fuses the multi-source battlefield environment data from different sensors and saves them in the environment database.

[0109] Step 2: The battlefield situation analysis module of the management decision-making layer performs battlefield situation analysis based on the battlefield environment data in the environment database, the firepower resource data in the firepower resource database, and the logistics resource data in the logistics database. The decision suggestion module generates a multi-level decision tree according to the battlefield situation analysis process. The plan generation module of the command intelligent body generates a preliminary resource management plan based on the battlefield situation analysis results and the multi-level decision tree.

[0110] Step 3: The command and coordination module of the command agent generates corresponding instructions according to the resource management plan and sends the instructions to the firepower dispatch module of the firepower agent and the logistics dispatch module of the support agent to execute the resource management plan and perform combat resource management.

[0111] Step 4. The real-time data update module inputs the real-time battlefield environment data into the environment database. The real-time monitoring module of the optimization and adjustment layer monitors the battlefield environment data in real time. The program evaluation module evaluates the applicability of the current resource management program based on the applicability evaluation index system with the real-time battlefield environment data. When the applicability score is less than 80 points, the battlefield situation analysis module and the real-time monitoring module of the management decision-making layer are reactivated to regenerate the optimized multi-level decision tree. The command agent regenerates the resource management plan based on the optimized multi-level decision tree, and sends it to the firepower scheduling module of the firepower agent and the logistics scheduling module of the support agent for plan execution.

[0112] Preferably, the step 2 of the present invention specifically comprises the following steps:

[0113] Step 21: The battlefield situation analysis module conducts threat assessment by fusing data, calculates the threat index of enemy units to our defense line or resources, and sets a threat index threshold. If the threat index of a certain area is higher than the threshold, an early warning will be triggered, and resources will be concentrated in that area first.

[0114] Step 22: The battlefield situation analysis module calls the firepower resource data in the firepower resource database and the logistics resource data in the logistics database to check the distribution and availability of firepower resources, including the current status of various types of weapon resources, remaining ammunition, and geographical location, and conducts a comprehensive inventory of logistics resources, including the quantity of supplies, transportation routes, and expected replenishment time.

[0115] Step 23: Use multi-source fusion data to perform battlefield environment and battlefield situation analysis, visualize threat information and resource availability in each area, so that the command agent can quickly understand the current situation. Based on the battlefield environment and battlefield situation analysis, the decision recommendation module generates a multi-level decision tree.

[0116] Preferably, the multi-level decision tree includes a plurality of nodes, each node in the multi-level decision tree represents a specific tactical choice, and the optimal decision is obtained by analyzing the results of each option. For example, in the multi-level decision tree, areas with high threat indexes can be prioritized to concentrate firepower, while logistics resources can be dispatched to support other areas.

[0117] Step 24: The plan generation module of the command agent generates a preliminary resource management plan based on the multi-level decision tree. It is recommended that the firepower agent concentrate firepower to attack high-threat areas, while commanding the logistics agent to provide material support to adjacent areas.

[0118] Based on the real-time feedback data from the environment simulation layer, the decision-making recommendation module can refine the plan. For example, if the threat index of a certain area increases, the system can automatically dispatch additional resources to cope with the change. At the same time, the system will continuously optimize the resource management plan based on the feedback of each intelligent agent to make the plan more accurate and effective.

[0119] Preferably, the step 4 of the present invention specifically comprises the following steps:

[0120] Step 41: Collect the latest battlefield data through the sensor module of the reconnaissance agent and update it to the environment database in real time.

[0121] Step 42: The solution evaluation module calls the applicability evaluation system according to the latest data and calculates the applicability of the current resource management solution through scoring.

[0122] Step 43: If the score is lower than the threshold, the system will start the real-time monitoring module, which will transmit the latest data to the solution generation module of the command agent to generate a new resource management solution.

[0123] Step 44: Send the new resource management plan to the command agent, which will issue optimized resource scheduling instructions and coordinate the firepower agent and the logistics agent to execute according to the optimized plan.

[0124] After completing the task, each agent will provide real-time execution feedback. The command agent will continuously monitor the execution effect of the plan based on these feedbacks and fine-tune the plan according to the battlefield dynamics to ensure that the resource management plan remains highly applicable in a dynamic battlefield environment.

[0125] Preferably, in the process of generating the optimization plan, the system first collects the latest battlefield dynamic environment data and obtains the execution feedback of the current resource management plan. Through data analysis, the bottleneck problems of the current plan are identified, such as insufficient allocation of firepower resources or delayed logistics supply. Based on these problems, the real-time monitoring module formulates corresponding optimization strategies, including adjusting the allocation of firepower resources, resetting the priority of logistics supply, and using the path optimization algorithm to adjust the transportation route to ensure that the materials can reach the front line more quickly. Subsequently, the system generates a new resource management plan, including the optimized firepower strike plan and logistics supply plan, and verifies the applicability of the plan through simulation. If the verification result shows that the plan meets the applicability standard, the system sends the optimization plan to the command intelligent body to issue an execution instruction; if the verification fails, it returns to the optimization process for further adjustment until the plan passes the verification.

[0126] The present invention provides a simulated combat resource management system and method based on a multi-agent system. Through the collaboration and autonomy of the multi-agent system, real-time dynamic management of combat resources is achieved to improve resource utilization efficiency. Through multi-source data fusion, accurate decision support is provided, resource allocation is optimized, and combat effectiveness is improved. Through a multi-level decision tree and a suitability evaluation index system, the system can provide accurate decision support information, optimize resource allocation, and improve combat effectiveness in a complex and changeable battlefield environment.

[0127] The above content is an explanation of the preferred embodiments of the present invention, which can help those skilled in the art to more fully understand the technical solution of the present invention. However, these embodiments are merely illustrative, and it cannot be determined that the specific implementation methods of the present invention are limited to the description of these embodiments. For ordinary technicians in the technical field to which the present invention belongs, without departing from the concept of the present invention, several simple deductions and transformations can be made, which should be regarded as belonging to the protection scope of the present invention.

Claims

1. A simulated combat resource management system based on a multi-agent system, characterized in that: It includes a multi-agent system layer, an environmental simulation layer, a management decision layer and an optimization adjustment layer; the multi-agent system layer includes a command agent, a communication agent, a reconnaissance agent, a fire agent and a support agent interconnected through a communication network; the environmental simulation layer includes a multi-source data fusion module, a real-time data update module and an environmental database; the management decision layer includes a battlefield situation analysis module and a decision suggestion module; the optimization adjustment layer includes a program evaluation module, a real-time monitoring module and a data breakpoint module; The reconnaissance agent collects multi-source battlefield environment data through multiple sensors, and the multi-source data fusion module uses weighted average fusion to fuse the multi-source battlefield environment data acquired by the reconnaissance agent by assigning weights to battlefield environment data from different data sources, and generates battlefield fusion data by performing multi-source data fusion; The battlefield situation analysis module performs battlefield situation analysis based on battlefield fusion data, and the decision suggestion module is used to generate a multi-level decision tree based on the battlefield situation analysis results and send it to the command intelligent agent to generate a simulated combat resource management plan, and the firepower intelligent agent and the support intelligent agent execute the simulated combat resource management plan; The real-time monitoring module monitors the battlefield environment in real time according to the battlefield environment data in the environment database; the solution evaluation module evaluates the applicability of the current resource management solution according to the applicability evaluation index system with the real-time battlefield environment data, and determines whether the resource management solution needs to be optimized and adjusted; the data breakpoint module monitors the value of the storage address of the multi-source battlefield environment data collected by each sensor in the environment simulation layer when the system is running. When these values ​​change unexpectedly, the system will suspend execution at the address, and check the call stack and other context information at that time to determine the reason for the change in the address value, and send the reason to the command intelligent agent, which will formulate a recovery plan. After the recovery plan is completed, the suspension of execution is lifted; The data breakpoint module shares the monitoring results of the sensor data with the solution evaluation module. When the data breakpoint module detects unexpected changes, it notifies the solution evaluation module, which then re-evaluates the applicability of the current resource management solution and determines whether the resource management solution needs to be optimized and adjusted based on the information provided by the data breakpoint module.

2. According to claim 1, a simulated combat resource management system based on a multi-agent system is characterized in that: The command intelligent agent, the communication intelligent agent, the reconnaissance intelligent agent, the firepower intelligent agent and the support intelligent agent are mutually independent subsystems, adopt a distributed network architecture, and are distributed on different combat entities; The command agent includes a command and coordination module and a plan generation module. The command and coordination module commands and coordinates the actions of other agents, and the plan generation module generates a simulated combat resource management plan. The communication agent includes a data communication module and a network management module. The data communication module establishes and manages data transmission links with other agents, and the network management module supports self-repair and self-optimization of the network. The reconnaissance agent includes a sensor module, a target recognition module and a data processing module. The sensor module includes a sensor network composed of multiple sensors for sensing the battlefield environment and collecting battlefield environment data. The target recognition module identifies enemy targets and combat resources. The data processing module pre-processes the battlefield environment data collected by the sensor module. The firepower agent includes a firepower resource database and a firepower scheduling module. The firepower resource data stores and manages information of various firepower resources. The firepower scheduling module schedules firepower resources to perform strike missions. The support intelligent body includes a logistics resource database and a logistics scheduling module. The logistics resource database stores and manages information of various logistics resources. The logistics scheduling module schedules logistics resources to support combat operations.

3. The multi-agent system-based simulated combat resource management system according to claim 1, characterized in that: The process of the multi-source data fusion module fusing the multi-source battlefield environment data acquired by the reconnaissance agent includes: Divide the multi-source battlefield environment data collected by multiple sensors received by the sensor module of the reconnaissance agent into M data sources, where M is the number of different battlefield environment data types received by the sensor module, where M≥2; Divide the battlefield into N battlefield areas, namely, the first battlefield area, the second battlefield area, the third battlefield area, ..., the Nth battlefield area, where N ≥ 2; The battlefield environment data from multiple different data sources are numbered and then integrated into multi-source battlefield environment data D1, D2, D3, ..., D M , where D1, D2, D3, …, D M Represents battlefield environment data from different data sources respectively; battlefield environment data D of the Mth data source in multi-source battlefield environment data M =[D M1 , D M2 , D M3 , …, D MN ], where D M1 , D M2 , D M3 , …, D MN They are battlefield environment data of different battlefield areas in the Mth data source; Weights are assigned to multi-source battlefield environment data, and different weight values ​​W are assigned to different data sources, where the weight value W=[W1, W2, W3, ..., W M ], where W1, W2, W3, …, W M Represent the weights of data from different data sources, W1+W2+W3+…+W M =1; The multi-source battlefield data from different data sources are fused by weighted average fusion technology to obtain fused data D 融合 , the fused data D 融合 =[W1*D1,W2*D2,W3*D3…W M *D M ], where the data from different data sources and their weights correspond one to one.

4. The multi-agent system-based simulated combat resource management system according to claim 1, characterized in that: The multi-level decision tree includes an upper-level firepower resource decision tree and a lower-level logistics resource decision tree. The firepower resource decision tree includes three decision nodes: threat assessment, firepower resource sufficiency, and reinforcement firepower completion time; the logistics resource decision tree includes three decision nodes: logistics resource sufficiency, front-line demand, and path optimization.

5. The multi-agent system-based simulated combat resource management system according to claim 4, characterized in that: The firepower resource decision tree performs threat assessment through battlefield fusion data and calculates the threat index of enemy units to our targets. The calculation formula of the threat index is: T=α×d+β×v, where T is the threat index, the d parameter is the distance between the enemy unit and our target collected by the sensor module, the v parameter is the enemy movement speed collected by the sensor module, and α and β are parameter weights.

6. The multi-agent system-based simulated combat resource management system according to claim 1, characterized in that: The resource management plan includes a firepower resource management plan and a logistics resource scheduling plan; the firepower resource management plan includes a target area sub-plan, a weapon type sub-plan, and a strike intensity and frequency sub-plan; the logistics resource scheduling plan includes a priority supply area sub-plan, a material type and quantity sub-plan, and a transportation route sub-plan.

7. The multi-agent system-based simulated combat resource management system according to claim 1, characterized in that: The applicability evaluation index system includes three levels: overall goals, first-level indicators and second-level indicators; the overall goal of the applicability evaluation index system is applicability scoring, which reflects the overall effectiveness of the resource management plan; the first-level indicators include five first-level indicator items: task achievement indicators, resource utilization indicators, battlefield control indicators, adaptability and flexibility indicators, and sustainability indicators; the task achievement indicators include three second-level indicator items: firepower strike accuracy indicator, logistics supply punctuality indicator, and enemy containment rate indicator; the resource utilization indicators include three second-level indicator items: firepower resource utilization indicator, logistics resource utilization indicator, and resource scheduling error rate indicator; the battlefield control indicator includes three second-level indicator items: enemy containment rate indicator, key zone control rate indicator, and enemy advancement suppression rate indicator; the adaptability and flexibility indicators include three second-level indicator items: material demand adaptability indicator, enemy dynamic adaptability indicator, and real-time feedback adjustment capability indicator; the adaptability and flexibility indicators include two second-level indicator items: battlefield persistence indicator and resource continuous supply capability indicator.

8. The multi-agent system-based simulated combat resource management system according to claim 7, characterized in that: The secondary indicator score is calculated through actual execution effect and plan achievement rate; The calculation formula for the applicability score of the first-level indicator is: , where m is the total number of secondary index items, 1≤n≤m, is the secondary indicator score for the corresponding secondary indicator item, is the weight of each secondary indicator item; The calculation formula for the overall target suitability score is: , Where j is the total number of first-level index items, 1≤k≤j, is the first-level indicator score for the corresponding first-level indicator item, is the weight of each first-level indicator in the overall applicability.

9. A method for managing simulated combat resources based on a multi-agent system, used to implement the simulated combat resource management system based on a multi-agent system as claimed in any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: The sensor module of the reconnaissance agent collects multi-source battlefield environment data collected by multiple sensors, and transmits the data to the environment simulation layer through the data communication module of the communication agent unit. The multi-source data fusion module of the environment simulation layer fuses the multi-source battlefield environment data from different sensors and saves them in the environment database; Step 2: The battlefield situation analysis module of the management decision-making layer performs battlefield situation analysis based on the battlefield environment data in the environment database, the firepower resource data in the firepower resource database, and the logistics resource data in the logistics database. The decision suggestion module generates a multi-level decision tree based on the battlefield situation analysis. The solution generation module of the command agent generates a preliminary resource management solution based on the battlefield situation analysis results and the multi-level decision tree. Step 3: The command and coordination module of the command agent generates corresponding instructions according to the resource management plan and sends the instructions to the firepower dispatch module of the firepower agent and the logistics dispatch module of the support agent to execute the resource management plan and perform combat resource management; Step 4. The real-time data update module inputs the real-time battlefield environment data into the environment database. The real-time monitoring module of the optimization and adjustment layer monitors the battlefield environment data in real time. The program evaluation module evaluates the applicability of the current resource management program based on the applicability evaluation index system with the real-time battlefield environment data. When the applicability score is less than 80 points, the battlefield situation analysis module and the real-time monitoring module of the management decision-making layer are reactivated to regenerate the optimized multi-level decision tree. The command agent regenerates the resource management plan based on the optimized multi-level decision tree, and sends it to the firepower scheduling module of the firepower agent and the logistics scheduling module of the support agent for plan execution.

Images